Note 7 - Risk Management

12 Months Ended

Dec. 31, 2020

Risk Management Abstract

Risk Management Explanatory

7. Risk management

7.1. Risk factors

BBVA Group has processes in place for identifying risks and analyzing scenarios in order to enable the Group to manage risks in a dynamic and proactive way.

The risk identification processes are forward looking to seek the identification of emerging risks and take into account the concerns of both the business areas, which are close to the reality of the different geographical areas, and the corporate areas and senior management.

Risks are identified and measured consistently using the methodologies deemed appropriate in each case. Their measurement includes the design and application of scenario analyses and stress testing and considers the controls to which the risks are subjected. As part of this process, a forward projection of the risk appetite framework variables in stress scenarios is conducted in order to identify possible deviations from the established thresholds. If any such deviations are detected, appropriate measures are taken to keep the variables within the target risk profile.

In this context, there are a number of emerging risks that could affect the evolution of the Group's business. These risks are included in the following blocks:

The coronavirus (COVID-19) pandemic is adversely affecting the world economy and economic activity and conditions in the countries in which the Group operates, leading many of them to economic recession in 2020 and relatively moderate activity growth in 2021, so that probably only from 2022 will the GDP levels observed before the crisis recover. Among other challenges, these countries are experiencing widespread increases in unemployment levels and falls in production, while public debt has increased significantly due to support and spending measures implemented by government authorities. In addition, there is an increase in defaults on debts by both companies and individuals, volatility in financial markets, including exchange rates, and falls in the value of assets and investments, all of which have had a negative impact on the Group's in the year 2020 and is expected to continue to affect in the future.

Furthermore, the Group may be affected by the measures adopted by regulatory authorities in the banking sector, including but not limited to, the recent reductions in reference interest rates, the relaxation of prudential requirements, the suspension of dividend payments, the adoption of payment deferrals measures for bank clients (such as those included in Royal Decree Law 11/2020 in Spain, as well as in the CECA-AEB agreement to which BBVA has adhered and which, among other things, allows loan debtors to extend maturities and defer interest payments) and facilities to grant credit through a line of guarantees or public guarantees, especially to companies and the self-employed individuals, as well as any changes in the financial asset purchase programs.

Since the outbreak of COVID-19 pandemic, the Group has experienced a decline in its activity. For example, the granting of new loans to individuals has significantly decreased since the beginning of mobility restriction measures approved in certain countries in which the Group operates. In addition, the Group faces several risks, such as a greater risk of impairment of the value of its assets (including financial instruments valued at fair value, which may undergo significant fluctuations) and of the securities held for liquidity reasons, a possible significant increase in non-performing loans and a negative impact on the cost of the Group's financing and its access to financing (especially in an environment where credit ratings are affected).

Furthermore, in several of the countries in which the Group operates, including Spain, the Group has temporarily closed a significant number of its offices and reduced opening hours to the public, and the teams that provide central services have been working remotely. Although these measures have been gradually reversed due to the continued expansion of the COVID-19 pandemic, it is unclear how long it will take until normal operations can fully resume. On the other hand, the pandemic could adversely affect the business and operations of third parties that provide critical services to the Group and, in particular, the higher demand and / or lower availability of certain resources could in some cases make it more difficult to maintain the service levels. In addition, the generalization of remote work has increased the risks related to cybersecurity, as the use of non-corporate networks has increased.

As a result of the above, the COVID-19 pandemic has had an adverse effect on the Group's results and capital base. During the first half of the year the main accumulated impacts were:

(i)an increase in the cost of risk associated with the lending activity, mainly due to the deterioration of the macroeconomic environment, which has had a negative impact of €2,009 million in the Group (including the initial adverse effect of the payment deferral) and provisions for credit risk and contingent commitments for €95 million, (see Notes 7.2, 46 and 47); and

(ii)a deterioration in the goodwill of the Group's subsidiary in the United States, mainly due to the deterioration of the macroeconomic scenario in the United States, which has had a net negative impact of €2,084 million on the Group's attributed profit in this period (although this impact does not affect the tangible book value, nor the solvency or the liquidity of the Group) (see Notes 18.1 and 49).

From June 30, 2020 on, and as a consequence of the general deterioration of the global macroeconomic scenario, its specific effects cannot be isolated, affecting all of the Group's consolidated Financial Statements.

Macroeconomic and geopolitical risks

The Global economy is being severely affected by the COVID-19 pandemic. Supply, demand and financial factors caused an unprecedented fall in GDP in the first half of 2020. Supported by strong fiscal and monetary policy measures, as well as greater control over the spread of the virus, global growth rebounded more than expected in the third quarter, before slowing down in the fourth, when the number of infections rose again in many regions, mainly in the United States and Europe. As for 2021, the unfavorable evolution of the pandemic is expected to adversely affect activity in the short term, while new fiscal and monetary stimuli, as well as the administering of coronavirus vaccines, are expected to support recovery from mid-year onwards.

Following the massive fiscal and monetary stimuli to support economic activity and reduce financial tensions, government debt has increased across the board and interest rates have been cut, and are now at historical low levels. Additional countercyclical measures may be required. Similarly, a significant reduction in current stimuli is not expected, at least until the recovery takes hold.

Tensions in the financial markets have moderated rapidly since the end of March 2020, following the decisive actions taken by the main central banks and the fiscal packages announced in many countries. In recent months, the markets have shown relative stability and, at certain times, risk-taking movements. Likewise, progress related to the development of COVID-19 vaccines and prospects for economic recovery should pave the way for financial volatility to persist at relatively low levels in general going forward.

BBVA Research estimates that global GDP contracted by around 2.6% in 2020 and will expand by around 5.3% in 2021 and 4.1% in 2022. Activity will recover gradually and heterogeneously among countries. Various epidemiological, financial and geopolitical factors are also contributing to the persistent exceptionally high uncertainty.

With regard to the banking system, in an environment in which much of the economic activity has been at a stand still for several months, the services provided have played an essential role, basically for two reasons: firstly, the banks have ensured the proper functioning of collections and payments for households and companies, thereby contributing to the maintenance of economic activity; secondly, the granting of new lending or the renewal of existing lending has reduced the impact of the economic slowdown on household and business income. The support provided by the banks over the months of lockdown and public guarantees have been essential in softening the impact of the crisis on companies' liquidity and solvency, meaning that banking has become its main source of funding for most companies.

In terms of profitability, European and Spanish banking have deteriorated, primarily because many entities recorded high provisions for impairment on financial assets in the first two quarters of 2020 as a result of the worsening macroeconomic environment following the pandemic outbreak. Pre-pandemic profitability levels remained far from the levels prior to the previous financial crisis. This is in addition to the accumulation of capital since the previous crisis and the very low interest rate environment that we have been experiencing for several years. Nevertheless, the banks are facing this situation from a healthy position and with solvency that has been constantly increasing since the 2008 crisis, with reinforced capital and liquidity buffers and, therefore, with a greater lending capacity.

The BBVA Group has a General Risk Management and Control Model appropriate to its business model, its organization, the countries in which it operates and its corporate governance system, which allows it to carry out its activity within the framework of the risk management and control strategy and policy defined by the corporate bodies. This model deals with management in global form adapting itself to the circumstances of each moment. This Model is applied integrally in the Group.

In this sense, from the beginning of the crisis, the BBVA Group implemented specific measures for the proper management of these associated risks, establishing different global initiatives that define the risk management strategy during the crisis, with common action protocols that should be implemented and adapted, when needed, to local needs.

The BBVA Group global risk unit - Global Risk Management (hereinafter, “GRM”) - has increased the frequency and intensity of the evaluation of potential impacts on the different groups and clients, in order to prevent their future evolution, and carried out appropriate adjustments and reclassifications, reinforcing its processes, governance and teams in Holding and countries to act in a coordinated manner, focusing priority on crisis management

Over the past year, it has been found that the pandemic has a global impact, affecting to a greater extent the sectors in which there is a high level of human interaction (transport, especially air transport, leisure, especially hotel establishments, as well as industries and activities dependent on them), regardless of the regional area in question. For this reason, the Bank's risk management has clearly been intensified by sectorial vectors over other conditioning factors such as geographic.

Regulatory and reputational risks

Financial institutions are exposed to a complex and ever-changing regulatory environment defined by governments and regulators. This can affect their ability to grow and the capacity of certain businesses to develop, and result in stricter liquidity and capital requirements with lower profitability ratios. The Group constantly monitors changes in the regulatory framework that allow for anticipation and adaptation to them in a timely manner, adopt industry practices and more efficient and rigorous criteria in its implementation.

The financial sector is under ever closer scrutiny by regulators, governments and society itself. In the course of activities, situations which might cause relevant reputational damage to the entity could raise and might affect the regular course of business. The attitudes and behaviors of the Group and its members are governed by the principles of integrity, honesty, long-term vision and industry practices through, inter alia, the internal control model, the Code of Conduct, the Corporate Principles in tax matters and Responsible Business Strategy of the Group.

Business, operational and legal risks

New technologies and forms of customer relationships: Developments in the digital world and in information technologies pose significant challenges for financial institutions, entailing threats (new competitors, disintermediation, etc.) but also opportunities (new framework of relations with customers, greater ability to adapt to their needs, new products and distribution channels, etc.). Digital transformation is a priority for the Group as it aims to lead digital banking of the future as one of its objectives.

Technological risks and security breaches: The Group is exposed to new threats such as cyber-attacks, theft of internal and customer databases, fraud in payment systems, etc. that require major investments in security from both the technological and human point of view. The Group gives great importance to the active operational and technological risk management and control.

The financial sector faces an environment of increasing regulatory and litigious pressure, and thus, the various Group entities are usually party to individual or collective judicial proceedings (including class actions) resulting from their activity and operations, as well as arbitration proceedings. The Group is also party to other government procedures and investigations, such as those carried out by the antitrust authorities in certain countries which, among other things, have in the past and could in the future result into sanctions, as well as lead to claims by customers and others. In addition, the regulatory framework, in the jurisdictions in which the Group operates, is evolving towards a supervisory approach more focused on the opening of sanctioning proceedings while some regulators are focusing their attention on consumer protection and behavioral risk.

In Spain and in other jurisdictions where the Group operates, legal and regulatory actions and proceedings against financial institutions, prompted in part by certain judgments in favor of consumers handed down by national and supranational courts, have increased significantly in recent years and this trend could continue in the future. The legal and regulatory actions and proceedings faced by other financial institutions in relation to these and other matters, especially if such actions or proceedings result in favorable resolutions for the consumer, could also adversely affect the Group.

All of the above may result in a significant increase in operating and compliance costs or even a reduction of revenues, and it is possible that an adverse outcome in any proceedings (depending on the amount thereof, the penalties imposed or the procedural or management costs for the Group) could damage the Group's reputation, generate a knock-on effect or otherwise adversely affect the Group.

It is difficult to predict the outcome of legal and regulatory actions and proceedings, both those to which the Group is currently exposed and those that may arise in the future, including actions and proceedings relating to former Group subsidiaries or in respect of which the Group may have indemnification obligations, but such outcome could be significantly adverse to the Group. In addition, a decision in any matter, whether against the Group or against another credit entity facing similar claims as those faced by the Group, could give rise to other claims against the Group. In addition, these actions and proceedings attract resources from the Group and may occupy a great deal of attention on part of the Group's management and employees.

As of December 31, 2020, the Group had €612 million in provisions for the proceedings it is facing (included in the line "Provisions for litigation and pending tax cases" in the consolidated balance sheet) (see Note 25), of which €574 million correspond to legal contingencies and €38 million to tax related matters. However, the uncertainty arising from these proceedings (including those for which no provisions have been made, either because it is not possible to estimate them or for other reasons) makes it impossible to guarantee that the possible losses arising from these proceedings will not exceed, where applicable, the amounts that the Group currently has provisioned and, therefore, could affect the Group's consolidated results in a given period.

As a result of the above, legal and regulatory actions and proceedings currently faced by the Group or to which it may become subject in the future or otherwise affected by, individually or in the aggregate, if resolved in whole or in part adversely to the Group ́s interests, could have a material adverse effect on the Group’s business, financial condition and results of operations.

Spanish judicial authorities are investigating the activities of Centro Exclusivo de Negocios y Transacciones, S.L. (Cenyt). Such investigation includes the provision of services by Cenyt to the Bank. On 29th July, 2019, the Bank was named as an official suspect (investigado) in a criminal judicial investigation (Preliminary Proceeding No. 96/2017 – Piece No. 9, Central Investigating Court No. 6 of the National High Court) for alleged facts which could be constitutive of bribery, revelation of secrets and corruption. On February 3, 2020, the Bank was notified by the Central Investigating Court No. 6 of the National High Court of the order lifting the secrecy of the proceedings. Certain current and former officers and employees of the Group, as well as former directors have also been named as official suspects in connection with this investigation. The Bank has been and continues to proactively collaborate with the Spanish judicial authorities, including sharing with the courts the relevant information obtained in the internal investigation hired by the entity in 2019 to contribute to the clarification of the facts. As of the date of the approval of the consolidated financial statements, no formal accusation against the Bank has been made.

This criminal judicial proceeding is at the pre-trial phase. Therefore, it is not possible at this time to predict the scope or duration of such proceeding or any related proceeding or its or their possible outcomes or implications for the Group, including any fines, damages or harm to the Group’s reputation caused thereby.

7.2. Credit risk

Credit risk arises from the probability that one party to a financial instrument will fail to meet its contractual obligations for reasons of insolvency or inability to pay and cause a financial loss for the other party.

The general principles governing credit risk management in the BBVA Group are:

Risks taken should comply with the general risk policy established by the Board of Directors of BBVA.

Risks taken should be in line with the level of equity and generation of recurring revenue of the BBVA Group prioritizing risk diversification and avoiding relevant concentrations.

Risks taken should be identified, measured and assessed and there should be management and monitoring procedures, in addition to sound mitigation and control mechanisms.

Risksshould be managed in a prudent and integrated manner during their life cycle and their treatment should be based on the type of risk. In addition, portfolios should be actively managed on the basis of a common metric (economic capital).

The main criterion when granting credit risks is the capability of the borrower or obligor to fulfill on a timely basis all financial obligations with its business income or source of income without depending upon guarantors, bondsmen or pledged assets.

Improve the financial health of our clients, help them in their decision making and in the daily management of their finances based on personalized advice.

Help our clients in the transition towards a sustainable future, with a focus on climate change and inclusive and sustainable social development.

Credit risk management in the Group has an integrated structure for all its functions, allowing decisions to be taken objectively and independently throughout the life cycle of the risk.

At Group level: frameworks for action and standard rules of conduct are defined for handling risk, specifically, the channels, procedures, structure and supervision.

At the business area level: they are responsible for adapting the Group's criteria to the local realities of each geographical area and for direct management of risk according to the decision-making channel:

Retail risks: in general, the decisions are formalized according to the scoring tools, within the general framework for action of each business area, with regard to risks. The changes in weighting and variables of these tools must be validated by the GRM area.

Wholesale risks: in general, the decisions are formalized by each business area within its general framework for action with regard to risks, which incorporates the delegation rule and the Group's corporate policies.

The risk function has a decision-making process supported by a structure of committees with a solid governance scheme, which describes their purposes and functioning for a proper performance of their tasks.

Payment deferral

This governance scheme has been fundamental in managing the COVID-19 crisis in all the geographies where the Group operates, where both, assessing the flow of necessary funds for the economies with the rigor in the analysis and monitoring of the credit quality of the exposures.

Since the beginning of the pandemic, the Group has offered payment deferral to its customers (retail, SMEs and wholesale) in all the geographies where it operates. These moratoriums have been both legislative (based on national laws) and non-legislative (based on sectorial or individual schemes), aimed at mitigating the effects of COVID-19. Depending on the cases, the payment of principal and / or interest has been postponed, maintaining the original contract. Generally, these deferrals have been given for a period of less than one year. This measure has been extended to different sectors, being Leisure, Real Estate and Auto the main users. The deadline to qualify for the moratorium has been extended in some geographies in recent months, having already come to an end in Mexico and Argentina. In the others, where this measure is still in force, this period ends in the first quarter of 2021, except Turkey (in May 2021), Colombia (in July 2021) and the USA (in January 2022).

Specifically, the Group's participation in the following moratorium or public guarantee measures by geography stands out:

In Spain, payment deferral measures have been covered mainly by Royal Decree Law 8/2020 and 11/2020, as well as the agreement promoted by the Spanish Banking Association (hereinafter “AEB”) to which BBVA has adhered.

The moratoriums covered by the Royal Decree Law have been proposed to the especially vulnerable groups indicated in the regulation. These measures consist of payment deferral of three months of principal and interest. In addition, the possibility of customers joining sector agreements for the remaining term until the limit established has been offered that, once said legal moratorium has expired. By type of customer, they are aimed at individuals, individual or self-employed entrepreneurs, and by type of product, mortgage, personal loans or consumer loans.

The moratoriums granted under the sectorial agreement of the AEB are aimed at individuals for up to 12 months of capital deferral in the case of mortgage loans and up to 6 months in personal loans. Said sector agreement has been in force until September 29, 2020, but it has been extended until March 30, 2021, although the new conditions only provide for the payment deferral of capital in mortgages up to 9 months, remaining 6 months on personal loans.

In addition, the Official Credit Institute (ICO) has published several aid programs aimed at the self-employed, SMEs and companies, through which a guarantee of between 60% and 80% is granted for a period of up to 5 years to the new financing granted. The amount of the guarantee and its length depends on the size of the company and the type of product. The ICO has also subsidized individuals the amount of the rent up to 6 months in loans up to 6 years.

In Mexico, the National Banking Commission of Securities (“CNBV”) published official letters P285/2020 of March 26, 2020 and P293/2020 of April 15, 2020, allowing the granting of capital and interest payment deferral for a period 4 months extendable for additional 2 additional months. These measures were mainly used by individuals and companies, affecting mortgage loans, personal loans and consumer loans, including credit cards.

In the United States, payment deferral measures have been mainly encouraged by the CARES Act, signed on March 27, 2020, which includes a wide range of supporting measures for companies and individuals, as well as an interagency statement (Office of the Comptroller of the Monetary Board of Governors of the Federal Reserve System, Federal Deposit Insurance Corporation, Office of Consumer Financial Protection and National Administration of Credit Unions) of April 7, 2020.

In Turkey, in mid March the government announced a program to stimulate the economy (Economic Stability Shield) allowing banks to defer payments for 3 months, with the possibility of extending up to 6 months, which was accompanied by several communications of the Banking Regulation and Supervisory Agency (“BRSA”) in this regard. These supporting measures are granted to both individuals and companies.

Likewise, public support programs have been recognized guaranteeing up to 80% of loans granted to companies for a period of 1 year.

In Colombia, the binding legislation for payment deferral comes from the Financial Superintendency, specifically from its Circulars 07/2020 and 14/2020, as well as from Resolution No. 385. The payment deferral consists of the deferral of payments of principal and interest until 6 months. The term to avail themselves of them has been extended until July 2021.

In Peru, measures were approved through various official letters issued by the Superintendency of Banking and Insurance (“SBS”), allowing the deferral of principal and interest payments initially up to 6 months and then extended to 12, mainly to individuals, self-employed and small companies.

Additionally, there have been public support programs such as “Reactiva”, “Crecer” or “FAE” aimed at companies and micro-companies with guaranteed amounts that, depending on the program and the type of company, are in a range of between 60% and 98%.

In Argentina, payment deferral measures are based on state legislation such as Royal Decree 544/2020 or Decree 319/202, as well as various Central Bank regulations. Aimed at a broad group of customers, they facilitate deferral of capital and interest for up to 3 months.

Certain public support programs offering guarantees of up to 100% to micro-SMEs or individuals and up to 25% to other companies in financing for up to 1 year.

The amount of payment deferrals (existing and completed) and the financing granted with public guarantees given at a Group level, as well as the number of customers of both measures, as of December 31, 2020 are as follows:

Amount of payment deferral and financing with public guarantees as of December 31, 2020 (Millions of Euros)
	Payment deferral				Financing with public guarantees
	Existing	Completed	Total	Number of customers	Total	Number ofcustomers	Total Payment deferral and guarantees	(%) credit investment
Group	6.803	27.025	33.828	2.843.977	18.619	271.870	52.446	13,1%

The amount of payment deferrals (existing and completed) and financing granted with public guarantees given at a Group level, broken down by segment, as of December 31, 2020 are as follows:

Amount of payment deferral and financing with public guarantees as of December 31, 2020 (Millions of Euros)
	Payment deferral			Financing with public guarantees
	Existing	Completed	Total
Group	6.803	27.025	33.828	18.619
Customers	4.657	16.676	21.333	1.237
Of which: Mortgages	3.664	8.723	12.387	1
SMEs	1.031	5.056	6.087	11.373
Non-financial corporations	1.055	5.095	6.150	5.930
Other	60	198	258	79

Amount of payment deferral by stages as of December 31, 2020 (Millions of Euros)
	Stage 1	Stage 2	Stage 3	Total
Group	21.670	9.761	2.397	33.828
Customers	13.608	5.920	1.805	21.333
Of which: Mortgages	8.310	3.163	914	12.387
SMEs	4.326	1.461	299	6.087
Non-financial corporations	3.495	2.362	293	6.150
Other	240	17	-	258

The payment deferral measures for bank customers result in the temporary suspension, total or partial, of the contractual obligations with a deferral for a specific period of time. Considering that the payment deferrals granted in connection with COVID-19 provide temporary relief to debtors and that the economic value of the affected loans has not been significantly impacted, the payment deferral measures granted have not been considered substantial contractual modifications and, therefore, modified loans are accounted for as a continuation of the originals. When a payment deferral does not generate interest collection rights, a temporary loss of value is triggered for the transaction, which is calculated as the difference between the current value of the original and the modified cash flows, both discounted at the effective interest rate of the original transaction. The difference is recognized at the initial time in the income statement under the heading “Impairment or (reversal) of impairment on financial assets not measured at fair value through profit or loss or net gains by modification” in the balance sheet as a reduction of the asset value of the loans. From that point on, said correction accrues as net interest income at the original effective interest rate within the period of the payment deferral. Thus, at the end of the moratorium period, the impact on net attributed profit is basically neutral.As of December 31, 2020, the temporary value loss of the payment deferral included in the consolidated income statements amounted to €304 million, from which €300 million have already been recognized as a higher interest margin as of that date.

Regarding the classification of exposures according to their credit risk, the Group has maintained a rigorous application of IFRS 9 when granting the payment deferrals and has reinforced the procedures for monitoring credit risk both throughout the life of the transactions and at their maturity.

This means that the payment deferrals granting does not imply in itself an automatic trigger for a significant increase in risk and that the transactions subject to the payment deferrals are initially classified in the stage they had previously, unless, based on their risk profile, they should be classified in a worse stage. On the other hand, as evidence of payment has ceased to exist or has been reduced, the Group has introduced additional indicators or segmentations to identify the significant increase in credit risk that may have occurred in some transactions or a set of them and, where appropriate, it has been classified in Stage 2. Furthermore, the indications provided by the European Banking Authority (EBA) have been taken into account to not consider forbearance the payment deferrals that meet a series of requirements. All this without prejudice to maintaining its consideration as a forbearance if it was previously qualified as such or classifying the exposure in the corresponding stage previously stated.

On the other hand, the treatment planned for the payment deferrals that expire and may require additional support will be in accordance with the updated evaluation of the customer's credit quality and the characteristics of the solution granted. If applicable, they will be treated as Refinancing or Restructuring as described in Note 7.2.7 of the Financial Statements.

Regarding public support for the loans’ lending, it does not affect the evaluation of the significant increase in risk since it is valued through the credit quality of the instrument. However, in estimating the expected loss, the existence of the guarantor implies a possible reduction in the level of provisions necessary since, for the hedged part, the loss that would be incurred in the foreclosure of the guarantee is taken into account.

The public guarantees granted in the different geographies in which the Group operates have been considered as an integral part of the terms and conditions of the loans granted under the consideration that the guarantees are granted at the same time that the financing is granted to the client and in a way inseparable from it.

7.2.1. Measurement of Expected Credit Loss (ECL)

IFRS 9 requires determining the expected credit loss (ECL) of a financial instrument in a way that reflects an unbiased estimation removing any conservatism or optimism, including the time value of money and a forward-looking perspective (including the economic forecast), all this based on the information that is available at a certain point in time and that is reasonable and bearable with respect to future economic conditions.

Therefore, the recognition and measurement of expected credit losses is highly complex and involves the use of significant analysis and estimation including formulation and incorporation of forward-looking economic conditions into the ECL model.

The modeling of the ECL calculation is subject to a governance system that is common to the entire Group. Within this common framework, each geography makes the necessary adaptations to capture its particularities. The methodology, assumptions and observations used by each geography are reviewed annually, and after a validation and approval process, the outcome of this review is incorporated into the ECL calculations.

Risk parameters by homogeneous groups

Expected losses can be estimated both individually and collectively. Regarding the collective estimate, the instruments are distributed in homogeneous groups (segments) that share similar risk characteristics. Following the guidelines established by the Group for the development of models under IFRS 9, each geography performs the grouping based on the information available, its representativeness or relevance and compliance with the necessary statistical requirements.

Depending on the portfolio or the parameter being estimated, one risk driver or another will apply and different segments will reflect differences in PDs and LGDs. Thus, in each segment, changes in the level of credit risk will respond to the impact of changing conditions on the common range of credit risk drivers. The effect on the group’s credit risk in response to changes in forward-looking information will be considered as well. Macroeconomic modeling for each segment is carried out using some of the shared risk characteristics.

These segments share credit risk characteristics such that changes in credit risk in a part of the portfolio are not concealed by the performance of other parts of the portfolio. In that sense, the methodology developed for ECL estimation indicates the risk drivers that have to be taken into account for PD segmentation purposes, depending on whether the estimation is for retail or wholesale portfolios.

As an example of the variables that can be taken into consideration to determine the final models, the following stand out:

PD - Retail: Contractual residual maturity, credit risk scoring, type of product, days past due, forbearance, time on books, time to maturity, nationality of the debtor, sale channel, original term, indicator of credit card activity, percentage of initial drawn balance.

PD - Wholesale: Credit Risk Rating, type of product, watch-list level, forbearance (client), time to maturity, industry sector, updated balance (y/n), written off, grace period.

LGD – retail: credit Risk Scoring, segment, type of product, secured / unsecured, type of collateral, sales channel, nationality, business area, debtor’s commercial segment, forbearance (account) EAD (this risk driver could be correlated with the time on books or the LTV so, before including it, an assessment should be done in order to avoid a double counting effect), time on default of the account (for defaulted exposures), geographical location.

LGD - wholesale: credit Risk Rating, geographical location, segment, type of product, secured / Unsecured, type of collateral, business area, forbearance (client), debtor’s commercial segment time on default of the deal (for defaulted exposures).

CCF: wholesale/retail, percentage of initial drawn balance,debtor’s commercial segment,days past due, forbearance, credit limit activity, time on books.

In the BBVA Group, the expected losses calculated are based on the internal models developed for all the Group's portfolios, unless clients are subject to individualized estimates.

Exposures with other credit institutions, sovereign debt or with public administrations are characterized by a low number of defaults, so the Group's historical bases do not contain sufficiently representative information to build impairment models based on them. However, there are external sources of information that, based on broader observations, are capable of providing the necessary inputs to develop models of expected losses. Therefore, based on the rating assigned to these exposures and taking into account the inputs obtained from these sources, the calculations of expected losses are developed internally, including their projection based on the macroeconomic perspectives.

Individual estimation of expected credit losses

The Group periodically and individually reviews the situation and credit rating of its customers, regardless of their classification, taking into consideration the information deemed necessary to do so. It also has procedures in place within the risk management framework to identify the factors that may lead to increased risk and, consequently, to a greater need for provisions.

The monitoring model established by the Group consists of continuously monitoring the risks to which it is exposed, which guarantees their proper classification in the different categories of IFRS 9. The original analysis of the exposures is reviewed through the procedures for updating the rating tools (rating and scoring), which periodically review the financial situation of clients, influencing the classification by stages of exposures.

Within this credit risk management framework, the Group has procedures that guarantee the review, at least annually, of all its wholesale counterparties through the so-called financial programs, which include the current and proposed positioning of the Group with the customer in terms of credit risk. This review is based on a detailed analysis of the client's up-to-date financial situation, which is complemented by other information available in relation to individual perspectives on business performance, industry trends, macroeconomic prospects or other public data. As a result of this analysis, the preliminary rating of the client is obtained, which, after undergoing the internal procedure, can be revised down if deemed appropriate (for example, general economic environment or evolution of the sector). These factors in addition to the information that the client can provide are used to review the ratings even before the scheduled financial plan reviews are conducted if circumstances warrant.

Additionally, the Group has established procedures to identify wholesale customers in the internal Watch List category, which is defined as that risk in which, derived from an individualized credit analysis, an increase in credit risk is observed, either due to economic or financial difficulties or because they have suffered, or are expected to suffer, adverse situations in their environment, without meeting the criteria for classification as impaired risk. Under this procedure, all a customer's Watch List exposures are considered Stage 2 regardless of when they originated, if as a result of the analysis the customer is considered to have significantly increased risk.

Finally, the Group has so-called Workout Committees, both local and corporate, which analyze not only the situation and evolution of significant clients in Watch List and doubtful situations, but also those significant clients in which, without having still rated on Watch List, they may present some Stage 2 rated exposure for a quantitative reason (PD comparison from origination). This analysis is carried out in order to decide if, derived from this situation, all the client's exposures should be considered in the Watch List category, which would imply the migration of all the client's operations to Stage 2 regardless of the date on which they originated.

With this, the Group supports an individualized review of the credit quality of its wholesale counterparties, identifying the situations in which a change in the risk profile of these clients may have occurred and proceeding, where appropriate, to estimate individualized credit losses.Along with this review, the Group individually estimates the expected losses of those clients whose total exposure exceeds certain thresholds, including those that part of their operations may be classified in stage 1 and part in stage 2. In setting thresholds, each geography determines the minimum amount of a client's exposure whose expected losses must be estimated individually taking into account the following:

For clients with exposures in stage 3. The analysis of clients with total risk above this threshold implies analyzing at least 40% of the total risk of the wholesale portfolio in stage 3. Although the calibration of the threshold is done on the wholesale portfolio, clients of other portfolios must be analyzed if they exceed the threshold, staying in Stage 3.

For all other situations. The analysis of clients with total risk above this threshold implies analyzing at least 20% of the total risk of the Watch List wholesale portfolio. Although the threshold calibration is carried out on the exposure classified as Watch List, wholesale clients or clients belonging to other portfolios that have exposures classified in stage 2 and whose total exposure exceeds the mentioned threshold must be analyzed individually, considering both the exposures classified in stage 1 as in stage 2.

Regarding the methodology for the individual estimation of expected losses, it should be mentioned, firstly, that these are measured as the difference between the asset’s carrying amount and the estimated future cash flows discounted at the financial asset’s effective interest

The estimated recoverable amount should correspond to the amount calculated under the following method:

The present value of estimated future cash flows discounted at the financial asset’s original effective interest rate; and

The estimation of the recoverable amount of a collateralized exposure reflects the cash flows that may result from the liquidation of the collateral.

The estimated future cash flows depend on the type of approach applied, which can be:

Going concern scenario: when the entity has updated and reliable information about the solvency and ability of payment of the holders or guarantors. The operating cash flows of the debtor, or the guarantor, continue and can be used to repay the financial debt to all creditors. In addition, collateral may be exercised to the extent it does not influence operating cash flows. The following aspects should be taken into account:

Future operating cash flows should be based on the financial statements of the debtor.

When the projections made on these financial statements assume a growth rate, a constant or decreasing growth rate must be used over a maximum growth period of 3 to 5 years, and subsequently constant cash flows

The growth rate should be based on the analysis of the evolution of the debtor's financial statements or on a sound and applicable business restructuring plan, taking into account the resulting changes in the structure of the company (for example, due to divestments or the interruption of unprofitable lines of business).

(Re)-investments that are needed to preserve cash flows should be considered, as well as any foreseeable future cash-flow changes (e.g. if a patent or a long-term loan expires).

When the recoverability of the exposure relies on the realization of the disposal of some assets by the debtor, the selling price should reflect the estimated future cash flows that may result from the sale of the assets less the estimated costs associated with the disposal.

Gone concern scenario: when the entity does not have updated and reliable information, it should consider that the estimation of loan receivable flows is of high uncertainty. Estimation should be carried out through the estimation of recoverable amounts from the effective real guarantees received. It will not be admissible as effective guarantees, those whose effectiveness depends substantially on the creditworthiness of the debtor or economic group in which it takes part. Under a gone concern scenario, the collateral is exercised and the operating cash flows of the debtor cease. This could be the case if:

The exposure has been past due for a long period. There is a rebuttable presumption that the allowance should be estimated under a gone concern criterion when arrears are greater than 18 months.

Future operating cash flows of the debtor are estimated to be low or negative.

Exposure is significantly collateralized, and this collateral is central to cash-flow generation

 There is a significant degree of uncertainty surrounding the estimation of the future cash flows. This would be the case if the earnings before interest, taxes, depreciation and amortization (EBITDA) of the two previous years had been negative, or if the business plans of the previous years had been flawed (due to material discrepancies in the back-testing)

 Insufficient information is available to perform a going concern analysis

Significant increase in credit risk

As indicated in Note 2.2, the criteria for identifying the significant increase in risk are applied consistently throughout the Group, distinguishing between quantitative reasons or by comparison of probabilities of default and qualitative reasons (more than 30 days of default, watch list consideration or non-impaired refinancing).

To manage credit risk, the Group uses all relevant information that is available and that may affect the credit quality of the exposures. This information may come mainly from the internal processes of admission, analysis and monitoring of operations, from the strategy defined by the Group regarding the price of operations or distribution by geographies, products or sectors of activity, from the observance of the macroeconomic environment, from market data such as interest rate curves, or prices of the different financial instruments, or from external sources of credit rating.

This set of information is the basis for determining the rating and scoring (see note 7.1.4 for more information on rating and scoring systems) corresponding to each of the exposures and which are assigned a probability of default (PD) that, as already mentioned, it undergoes an annual review process that assesses its representativeness (backtesting) and is updated with new observations. Furthermore, the projection of these PDs over time has been modeled based on macroeconomic expectations, which allows obtaining the probabilities of default throughout the life of the operations.

Based on this common methodology, and in accordance with the provisions of IFRS 9 and the EBA guidelines on credit risk management practices (EBA / GL / 2017/06), each geography has established absolute and relative thresholds for identifying whether the expected changes in the probabilities of default have increased significantly compared to the initial moment, adapted to the particularities of each one of them in terms of origination levels, product characteristics, distribution by sectors or portfolios, and macroeconomic situation. To establish the aforementioned thresholds, a series of general principles are considered, such as:

Uniformity: Based on the rating and scoring systems that, in a homogeneous manner, are implemented in the Group's units.

Stability: The thresholds must be established to identify the significant increase in risk produced in exposures since their initial recognition and not only to identify those situations in which it is already foreseeable that they will reach the level of impairment. For this reason, it is to be expected that of the total exposures there will always be a representative group for which said increased risk is identified.

Anticipation: The thresholds must consider the identification of the increased risk in advance with respect to the recognition of the exposures as impaired or even before a real default occurs. The calibration of the thresholds should minimize the cases in which the instruments are classified in stage 3 without having previously been recognized as stage 2.

Indicators or metrics: It is expected that the classification of the exposures in stage 2 will have sufficient permanence to allow them to develop an anticipatory management of them before, where appropriate, they end up migrating to stage 3.

Symmetry: IFRS 9 provides for a symmetric treatment both to identify the significant increase in risk and to identify that it has disappeared, so the thresholds also work to improve the credit classification of exposures. In this sense, it is expected that the cases in which the exhibitions that improve from stage 3 are directly classified into stage 1 will be minimal.

The identification of the significant increase in risk from the comparison of the probabilities of default should be the main reason why exposures in stage 2 are recognized.

Specifically, a contract will be transferred to stage 2 when the following two conditions are met by comparing the current PD values and the origination PD values:

These absolute and relative thresholds are consistently established for each geography and for each portfolio, taking into account their particularities and based on the principles described. The thresholds set by each geography are included within the annual review process and, generally, are in the range of 30% to 250% for the relative threshold and from 10 to 150 basis points for the absolute threshold.

The establishment of absolute and relative thresholds, as well as their different levels, comply with the provisions of IFRS 9 when it indicates that a certain change, in absolute terms, in the risk of a default will be more significant for a financial instrument with a lower initial risk of default compared to a financial instrument with higher initial risk of default.

For existing contracts before the implementation of IFRS 9, given the limitations in the information available on them, the thresholds are calibrated based on the PDs obtained from the prudential or economic models for calculating capital.

Risk Parameters Adjusted by Macroeconomic Scenarios

Expected Credit Loss must include forward looking information, in accordance with IFRS 9, which states that the comprehensive credit risk information must incorporate not only historical information but also all relevant credit information, including forward-looking macroeconomic information. BBVA Group uses the classical credit risk parameters PD, LGD and EAD in order to calculate the ECL for the credit portfolios.

BBVA Group´s methodological approach in order to incorporate the forward looking information aims to determine the relation between macroeconomic variables and risk parameters following three main steps:

Step 1: Analysis and transformation of time series data.

Step 2: For each dependent variable find conditional forecasting models that are economically consistent.

Step 3: Select the best conditional forecasting model from the set of candidates defined in Step 2, based on their forecasting capacity.

How economic scenarios are reflected in calculation of ECL

The forward looking component is added to the calculation of the ECL through the introduction of macroeconomic scenarios as an input. Inputs highly depend on the particular combination of region and portfolio, so inputs are adapted to available data regarding each of them.

Based on economic theory and analysis, the main indicators most directly relevant for explaining and forecasting the selected risk parameters (PD, LGD and EAD) are:

The net income of families, corporates or public administrations.

The outstanding payment amounts on the principal and interest on the financial instruments.

The value of the collateral assets pledge to the loan.

BBVA Group approximates these variables by using a proxy indicator from the set included in the macroeconomic scenarios provided by the economic research department.

Only a single specific indicator for each of the three categories can be used and only one of the following core macroeconomic indicators should be chosen as first option:

The real GDP growth for the purpose of conditional forecasting can be seen as the only “factor” required for capturing the influence of all potentially relevant macro-financial scenarios on internal PDs and LGD.

The most representative short term interest rate (typically the policy rate or the most liquid sovereign yield or interbank rate) or exchange rates expressed in real terms.

A comprehensive and representative index of the price of real estate properties expressed in real terms in the case of mortgage loans and a representative and real term index of the price of the relevant commodity for corporate loan portfolios concentrated in exporters or producer of such commodity.

Real GDP growth is given priority over any other indicator not only because it is the most comprehensive indicator of income and economic activity but also because it is the central variable in the generation of macroeconomic scenarios.

Multiple scenario approach under IFRS 9

IFRS 9 requires calculating an unbiased probability weighted measurement of expected credit losses (“ECL”) by evaluating a range of possible outcomes, including forecasts of future economic conditions.

The BBVA Research teams within the BBVA Group produce forecasts of the macroeconomic variables under the baseline scenario, which are used in the rest of the related processes of the Group, such as budgeting, ICAAP and risk appetite framework, stress testing, etc.

Additionally, the BBVA Research teams produced alternative scenarios to the baseline scenario so as to meet the requirements under the IFRS 9 standard.

Alternative macroeconomic scenarios

For each of the macro-financial variables, BBVA Research produces three scenarios.

BBVA Research tracks, analyzes and forecasts the economic environment to provide a consistent forward looking assessment about the most likely scenario and risks that impact BBVA’s footprint. To build economic scenarios, BBVA Research combines official data, econometric techniques and expert knowledge.

Each of these scenarios corresponds to the expected value of a different area of the probabilistic distribution of the possible projections of the economic variables.

The non-linearity overlay is defined as the ratio between the probability-weighted ECL under the alternative scenarios and the baseline scenario, where the scenario’s probability depends on the distance of the alternative scenarios from the base one.

BBVA Group establishes equally weighted scenarios, being the probability 34% for the baseline scenario, 33% for the worst alternative scenario and 33% for the best alternative scenario.

The approach in BBVA Group consists on using the scenario that is the most likely scenario, which is the baseline scenario, consistent with the rest of internal processes (ICAAP, Budgeting…) and then applying an overlay adjustment that is calculated by taking into account the weighted average of the ECL determined by each of the scenarios. This effect is calculated taking into account the weighted weight of the expected loss determined for each scenario.

In general, it is expected that the effect of the overlay is to increase the ECL. It is possible to obtain an overlay that does not have that effect, whenever the relationship between macro scenarios and losses is linear. However, the overlay is not expected to reduce the ECL On the other hand, the BBVA Group also takes into account the range of possible scenarios when defining its significant increase in credit risk. Thus, the PDs used in the quantitative process to identify the significant increase in credit risk will be those that result from making a weighted average of the PDs calculated under the three scenarios.

Macroeconomic scenarios as a result of the COVID-19 pandemic

The COVID-19 pandemic has generated a macroeconomic uncertainty situation with a direct impact on credit risk of the entities, particularly, on the expected credit losses under IFRS 9. Even though the situation is unclear and of an unforeseeable time length, the expectation is that this situation will provoke a severe recession followed by an economic recovery, which will not achieve the pre-crisis GDP levels in the short-term, supported by the measures issued by governments and monetary authorities.

This situation has allowed the accounting authorities and the banking supervisors to adopt measures in order to mitigate the impacts that this crisis could imply on the calculation of expected credit losses under IFRS 9 as well as on solvency, urging:

the entities to evaluate all the available information, weighing more the long-term forecasts against the short-term economic situation

the governments to adopt measures to avoid the effects of impairment,

the entities to develop managerial measures as the design of specific products adapted to the situation which could occur during this crisis.

Almost all accounting and prudential authorities have issued recommendations or measures within the COVID-19 crisis framework regarding the estimation of the expected losses under IFRS 9 in a coordinated manner.

The common denominator of all of these recommendations is that, given the difficulty of establishing reliable macroeconomic forecasts, the transitory term of the economic shock and the need to incorporate the effect of the mitigating measures issued by the governments, a review of the automatic application of the models in order to increase the weight of the long-term macroeconomic forecasts in the calculation of the expected losses is needed. As a result thereof, the expected outcome over the lifetime of the transactions will have more weight than the short-term macroeconomic impact.

In this respect, the BBVA Group has taken into account those recommendations in the calculation of the expected credit losses under IFRS 9, considering that the economic situation caused by the COVID-19 pandemic is transitory and will be followed by a recovery, even if there is uncertainty over the level and the time period of such recovery. As a consequence, different scenarios have been taken into consideration in the calculation of expected losses, resulting in the model management believes suits best the current economic situation and the combined recommendations issued by the authorities. In addition to the outcome of the calculation of the scenarios, individual analysis of exposures which could be most affected by the circumstances caused by the COVID-19, have been taken into account.

The estimate for the next five years of the Gross Domestic Product (GDP), of the variation in the unemployment rate and of the House Price Index (HPI), for the most relevant countries where it represents a significant factor, is determined by BBVA Research and it has been used at the time of the calculation of the expected credit loss as of December 31, 2020:

Positive scenario of GDP, unemployment rate and HPI for the main geographies
	Spain			Mexico			Turkey
Date	GDP	Unemployment	HPI	GDP	Unemployment	HPI	GDP	Unemployment
2020	(11,20%)	16,44%	(1,44%)	(8,85%)	4,57%	1,71%	2,07%	13,45%
2021	6,63%	16,03%	(3,28%)	4,58%	5,40%	(1,23%)	9,08%	12,60%
2022	6,27%	12,72%	4,56%	3,80%	5,17%	0,32%	5,30%	11,58%
2023	2,95%	10,82%	5,79%	1,62%	5,04%	0,31%	4,13%	11,58%
2024	2,07%	9,58%	3,66%	1,47%	4,91%	1,01%	4,11%	11,19%
2025	2,01%	8,55%	3,57%	1,47%	4,76%	1,72%	4,10%	10,85%

	Peru		Argentina		Colombia
Date	GDP	Unemployment	GDP	Unemployment	GDP	Unemployment
2020	(11,74%)	12,75%	(10,64%)	13,60%	(6,80%)	18,14%
2021	12,56%	10,29%	9,95%	14,39%	6,80%	16,14%
2022	5,25%	10,00%	3,52%	11,88%	3,70%	14,53%
2023	3,68%	8,73%	2,08%	8,99%	3,15%	14,28%
2024	3,58%	7,23%	2,11%	7,69%	3,27%	12,49%
2025	3,35%	6,88%	2,14%	6,78%	3,60%	12,28%

Estimate of GDP, unemployment rate and HPI for the main geographies
	Spain			Mexico			Turkey
Date	GDP	Unemployment	HPI	GDP	Unemployment	HPI	GDP	Unemployment
2020	(11,48%)	16,95%	(1,98%)	(9,25%)	4,62%	1,81%	(0,01%)	13,98%
2021	5,99%	17,51%	(5,08%)	3,71%	5,57%	(1,32%)	5,52%	14,05%
2022	6,04%	14,35%	3,48%	3,53%	5,35%	0,15%	4,53%	12,58%
2023	2,93%	12,41%	5,44%	1,55%	5,19%	0,31%	4,01%	11,95%
2024	2,07%	11,14%	3,20%	1,45%	5,03%	1,02%	3,99%	11,38%
2025	2,01%	9,99%	3,12%	1,46%	4,88%	1,71%	3,98%	11,03%

	Peru		Argentina		Colombia
Date	GDP	Unemployment	GDP	Unemployment	GDP	Unemployment
2020	(13,04%)	12,80%	(13,00%)	13,98%	(7,51%)	18,23%
2021	10,05%	10,48%	5,54%	15,40%	5,48%	16,40%
2022	4,52%	10,23%	2,54%	12,80%	3,46%	14,83%
2023	3,69%	8,93%	1,98%	9,60%	3,15%	14,57%
2024	3,58%	7,41%	1,98%	8,18%	3,27%	12,78%
2025	3,35%	7,06%	2,01%	7,28%	3,60%	12,55%

Negative scenario of GDP, unemployment rate and HPI for the main geographies
	Spain			Mexico			Turkey
Date	GDP	Unemployment	HPI	GDP	Unemployment	HPI	GDP	Unemployment
2020	(11,76%)	17,44%	(2,60%)	(9,64%)	4,67%	1,89%	(2,10%)	14,49%
2021	5,37%	18,94%	(6,69%)	2,84%	5,75%	(1,48%)	1,75%	15,51%
2022	5,82%	15,92%	2,49%	3,25%	5,53%	(0,06%)	3,56%	13,64%
2023	2,88%	13,99%	4,94%	1,48%	5,34%	0,17%	3,92%	12,33%
2024	2,03%	12,70%	2,45%	1,41%	5,17%	0,99%	3,91%	11,56%
2025	1,97%	11,45%	2,36%	1,41%	5,02%	1,70%	3,91%	11,20%

	Peru		Argentina		Colombia
Date	GDP	Unemployment	GDP	Unemployment	GDP	Unemployment
2020	(14,33%)	12,85%	(15,28%)	14,34%	(8,25%)	18,31%
2021	7,53%	10,69%	0,89%	16,38%	4,16%	16,66%
2022	3,78%	10,48%	1,33%	13,69%	3,16%	15,10%
2023	3,69%	9,15%	1,86%	10,19%	3,15%	14,84%
2024	3,57%	7,62%	1,83%	8,63%	3,27%	13,04%
2025	3,35%	7,27%	1,86%	7,75%	3,60%	12,80%

The estimate for the next five years of the Gross Domestic Product (GDP), is determined by BBVA Research and it has been used at the time of the calculation of the expected credit loss as of December 31, 2019:

Sensitivity to macroeconomic scenarios

A sensitivity exercise has been carried out on the expected losses due to variations in the key hypotheses as they are the ones that introduce the greatest uncertainty in estimating such losses. As a first step, GDP and House Prices have been identified as the most relevant variables. These variables have been subjected to shocks of +/- 100 bps in their entire projection window. Independent sensitivities have been assessed, under the assumption of assigning a 100% probability to each determined scenario with these independent shocks. 

Variation in provisions is determined both by re-staging (that is: in worse scenarios due to the recognition of lifetime credit losses for additional operations that are transferred to stage 2 from stage1 where 12 months of losses are valued: or vice versa in improvement scenarios) as well as variations in the collective risk parameters (PD and LGD) of each financial instrument due to the changes defined in the macroeconomic forecasts of the scenario.

Expected loss variation
	BBVA Group						Spain			Mexico			Turkey
GDP	Total Portfolio	Retail	Mortgages	Wholesaler	Fixed income	Total Portfolio	Mortgages	Companies	Total Portfolio	Mortgages	Cards	Total Portfolio	Mortgages	Cards
-100pb	3,55%	3,47%	3,72%	3,91%	1,58%	3,72%	4,39%	3,96%	3,91%	2,20%	6,30%	1,56%	1,58%	1,62%
+100pb	(3,25%)	(3,14%)	(3,03%)	(3,69%)	(1,97%)	(3,32%)	(3,57%)	(3,53%)	(3,64%)	(2,07%)	(5,78%)	(1,47%)	(1,55%)	(1,47%)
Housing price
-100pb							5,41%	0,79%		3,13%
+100pb							(5,35%)	(0,77%)		(4,47%)

Additional adjustments to expect loss measurement

In addition to what is described on individualized and collective estimates of expected losses and macroeconomic estimates, the Group may supplement the expected losses if it deems it necessary to collect the effects that may not be included, either by considering risk drivers or by the incorporation of sectorial particularities or that may affect a set of operations or borrowers. These adjustments should be temporary, until the reasons that motivated them disappear or materialize.

For this reason, the expected losses have been supplemented with additional amounts that have been considered necessary to collect the particular characteristics of borrowers, sectors or portfolios and that may not be identified in the general process. Of the supplementary amounts recognized throughout the year, at the end of the year 2020, 244 million euros are pending allocation to specific borrowers, of which 223 million euros are located in Spain (mainly 57 million euros based on the volume of arrears pending maturity and whose behavior pattern is still subject to uncertainty, 127 million euros in sectors most affected by the pandemic and 40 million euros as a complement to the individualized analyzes) and 21 million euros in the United States in relation to the uncertainty of the Oil & Gas sector.

7.2.2. Credit risk exposure

In accordance with IFRS 7 “Financial instruments: Disclosures”, the BBVA Group’s credit risk exposure by headings in the balance sheetsas of December 31, 2020, 2019 and 2018 is provided below. It does not consider the loss allowances and the availability of collateral or other credit enhancements to enable compliance with payment obligations. The details are broken down by financial instruments and counterparties:

Maximum credit risk exposure (Millions of Euros)
	Notes	December 2020	Stage 1	Stage 2	Stage 3
Financial assets held for trading		68.075
Debt securities	10	23.970
Equity instruments	10	11.458
Loans and advances	10	32.647
Non-trading financial assets mandatorily at fair value through profit or loss		5.198
Loans and advances	11	709
Debt securities	11	356
Equity instruments	11	4.133
Financial assets designated at fair value through profit or loss	12	1.117
Derivatives (trading and hedging)		46.302
Financial assets at fair value through other comprehensive income		69.537
Debt securities		68.404	67.995	410	-
Equity instruments	13	1.100
Loans and advances to credit institutions	13	33	33	-	-
Financial assets at amortized cost		379.857	334.552	30.607	14.698
Loans and advances to central banks		6.229	6.229	-	-
Loans and advances to credit institutions		14.591	14.565	20	6
Loans and advances to customers		323.252	277.998	30.581	14.672
Debt securities		35.785	35.759	6	20
Total financial assets risk		570.084	-	-	-
Total loan commitments and financial guarantees		179.440	165.726	12.682	1.032
Loan commitments given	33	132.584	124.104	8.214	265
Financial guarantees given	33	10.665	9.208	1.168	290
Other commitments given	33	36.190	32.414	3.300	477
Total maximum credit exposure		749.524

Maximum credit risk exposure (Millions of Euros)
	Notes	December 2019	Stage 1	Stage 2	Stage 3
Financial assets held for trading		69.503
Debt securities	10	26.309
Equity instruments	10	8.892
Loans and advances	10	34.303
Non-trading financial assets mandatorily at fair value through profit or loss		5.557
Loans and advances	11	1.120
Debt securities	11	110
Equity instruments	11	4.327
Financial assets designated at fair value through profit or loss	12	1.214
Derivatives (trading and hedging)		39.462
Financial assets at fair value through other comprehensive income		61.293
Debt securities		58.841	58.590	250	-
Equity instruments	13	2.420
Loans and advances to credit institutions	13	33	33	-	-
Financial assets at amortized cost		451.640	402.024	33.624	15.993
Loans and advances to central banks		4.285	4.285	-	-
Loans and advances to credit institutions		13.664	13.500	158	6
Loans and advances to customers		394.763	345.449	33.360	15.954
Debt securities		38.930	38.790	106	33
Total financial assets risk		628.670
Total loan commitments and financial guarantees		181.116	169.663	10.452	1.001
Loan commitments given	33	130.923	123.707	6.945	270
Financial guarantees given	33	10.984	9.804	955	224
Other commitments given	33	39.209	36.151	2.552	506
Total maximum credit exposure		809.786

Maximum credit risk exposure (Millions of Euros)
	Notes	December 2018	Stage 1	Stage 2	Stage 3
Financial assets held for trading		59.581
Debt securities	10	25.577
Equity instruments	10	5.254
Loans and advances	10	28.750
Non-trading financial assets mandatorily at fair value through profit or loss		5.135
Loans and advances	11	1.803
Debt securities	11	237
Equity instruments	11	3.095
Financial assets designated at fair value through profit or loss	12	1.313
Derivatives (trading and hedging)		38.249
Financial assets at fair value through other comprehensive income		56.365
Debt securities		53.737	53.734	3	-
Equity instruments	13	2.595
Loans and advances to credit institutions	13	33	33	-	-
Financial assets at amortized cost		431.927	384.632	30.902	16.394
Loans and advances to central banks		3.947	3.947	-	-
Loans and advances to credit institutions		9.175	9.131	34	10
Loans and advances to customers		386.225	339.204	30.673	16.348
Debt securities		32.580	32.350	195	35
Total financial assets risk		592.571
Total loan commitments and financial guarantees		170.511	161.404	8.120	987
Loan commitments given	33	118.959	113.403	5.308	247
Financial guarantees given	33	16.454	14.902	1.220	332
Other commitments given	33	35.098	33.099	1.591	408
Total maximum credit exposure		763.082

The maximum credit exposure presented in the table above is determined by type of financial asset as explained below:

In the case of financial instruments recognized in the consolidated balance sheets, exposure to credit risk is considered equal to its carrying amount (not including loss allowances) with the only exception of trading and hedging derivatives.

The maximum credit risk exposure on financial commitments and guarantees granted is the maximum that the Group would be liable for if these guarantees were called in, or the higher amount pending to be disposed from the customer in the case of commitments.

The calculation of risk exposure for derivatives is based on the sum of two factors: the derivatives fair value and their potential risk (or "add-on").

The breakdown by geographical location and Stage of the maximum credit risk exposure, the accumulated allowances recorded and the carrying amount of the loans and advances to customers as of December 31, 2020, 2019 and 2018 is shown below:

December 2020 (Millions of Euros)
	Gross exposure				Accumulated allowances				Carrying amount
	Total	Stage 1	Stage 2	Stage 3	Total	Stage 1	Stage 2	Stage 3	Total	Stage 1	Stage 2	Stage 3
Spain (*)	195.983	171.397	16.387	8.199	(5.679)	(753)	(849)	(4.077)	190.304	170.644	15.538	4.122
Mexico	52.211	46.373	4.071	1.767	(2.211)	(685)	(442)	(1.083)	50.000	45.688	3.628	684
Turkey (**)	39.633	30.832	5.806	2.995	(2.338)	(246)	(535)	(1.557)	37.295	30.586	5.272	1.438
South America (***)	34.499	28.484	4.312	1.703	(1.870)	(320)	(460)	(1.090)	32.629	28.165	3.852	612
Others	925	912	5	8	(7)	(1)	-	(6)	918	911	4	2
Total (****)	323.252	277.998	30.581	14.672	(12.105)	(2.005)	(2.287)	(7.813)	311.147	275.993	28.294	6.860
Of which: individual					(2.611)	(10)	(479)	(2.122)
Of which: collective					(9.494)	(1.995)	(1.808)	(5.691)

(*) Spain includes all countries where BBVA, S.A. operates.

(**) Turkey includes all countries in which Garanti BBVA operates.

(***) In South America, BBVA Group operates mainly in Argentina, Colombia, Peru and Uruguay.

(****) The amount of the accumulated impairment includes the provisions recorded for credit risk over the remaining expected lifetime of purchased financial instruments. Those provisions were determined at the moment of the Purchase Price Allocation and were originated mainly in the acquisition of Catalunya Banc S.A. (as of December 31, 2020, the remaining balance was €363 million). These valuation adjustments are recognized in the consolidated income statement during the residual life of the operations or are applied to the value corrections when the losses materialize.

December 2019 (Millions of Euros)
	Gross exposure				Accumulated allowances				Carrying amount
	Total	Stage 1	Stage 2	Stage 3	Total	Stage 1	Stage 2	Stage 3	Total	Stage 1	Stage 2	Stage 3
Spain (*)	197.058	173.843	14.599	8.616	(5.311)	(712)	(661)	(3.939)	191.747	173.131	13.939	4.677
The United States	57.387	49.744	7.011	632	(688)	(165)	(342)	(182)	56.699	49.580	6.670	450
Mexico	60.099	54.748	3.873	1.478	(2.013)	(697)	(404)	(912)	58.087	54.052	3.469	566
Turkey (**)	43.113	34.536	5.127	3.451	(2.613)	(189)	(450)	(1.974)	40.500	34.347	4.677	1.477
South America (***)	36.265	31.754	2.742	1.769	(1.769)	(366)	(323)	(1.079)	34.497	31.388	2.419	690
Others	839	824	7	9	(8)	(1)	(1)	(6)	832	823	6	2
Total (****)	394.763	345.449	33.360	15.954	(12.402)	(2.129)	(2.181)	(8.093)	382.360	343.320	31.179	7.861
Of which: individual					(2.795)	(6)	(347)	(2.441)
Of which: collective					(9.608)	(2.123)	(1.834)	(5.652)

(*) Spain includes all countries where BBVA, S.A. operates.

(**) Turkey includes all countries in which Garanti BBVA operates.

(***) In South America, BBVA Group operates mainly in Argentina, Chile, Colombia, Peru, Uruguay and Venezuela.

(****) The amount of the accumulated impairment includes the provisions recorded for credit risk over the remaining expected lifetime of purchased financial instruments. Those provisions were determined at the moment of the Purchase Price Allocation and were originated mainly in the acquisition of Catalunya Banc S.A. (as of December 31, 2019 the remaining balance was €433 million). These valuation adjustments are recognized in the consolidated income statement during the residual life of the operations or are applied to the value corrections when the losses materialize.

December 2018 (Millions of Euros)
	Gross exposure				Accumulated allowances				Carrying amount
	Total	Stage 1	Stage 2	Stage 3	Total	Stage 1	Stage 2	Stage 3	Total	Stage 1	Stage 2	Stage 3
Spain (*)	195.447	172.599	12.827	10.021	(5.874)	(713)	(877)	(4.284)	189.574	171.886	11.951	5.737
The United States	57.321	50.665	5.923	733	(658)	(206)	(299)	(153)	56.663	50.459	5.624	580
Mexico	52.858	48.354	3.366	1.138	(1.750)	(640)	(373)	(737)	51.107	47.714	2.992	401
Turkey (**)	43.718	34.883	6.113	2.722	(2.241)	(171)	(591)	(1.479)	41.479	34.712	5.523	1.244
South America (***)	36.098	31.947	2.436	1.715	(1.656)	(338)	(234)	(1.084)	34.442	31.609	2.202	631
Others	783	756	8	19	(19)	-	(1)	(18)	763	755	7	1
Total (****)	386.225	339.204	30.673	16.348	(12.199)	(2.070)	(2.374)	(7.755)	374.027	337.134	28.299	8.593
Of which: individual					(3.333)	(3)	(504)	(2.826)
Of which: collective					(8.866)	(2.067)	(1.870)	(4.929)

(*) Spain includes all countries where BBVA, S.A. operates.

(**) Turkey includes all countries in which Garanti BBVA operates.

(***) In South America, BBVA Group operates mainly in Argentina, Chile, Colombia, Peru, Uruguay and Venezuela.

(****) The amount of the accumulated impairment includes the provisions recorded for credit risk over the remaining expected lifetime of purchased financial instruments. Those provisions were determined at the moment of the Purchase Price Allocation and were originated mainly in the acquisition of Catalunya Banc S.A. (as of December 31, 2018 the remaining balance was €540 million). These valuation adjustments are recognized in the consolidated income statement during the residual life of the operations or are applied to the value corrections when the losses materialize.

The breakdown by counterparty of the maximum credit risk exposure, the accumulated allowances recorded, as well as the carrying amount by stages of loans and advances to customers as of December 31, 2020, 2019 and 2018 is shown below:

December 2020 (Millions of Euros)
	Gross exposure				Accumulated allowances				Net amount
	Total	Stage 1	Stage 2	Stage 3	Total	Stage 1	Stage 2	Stage 3	Total	Stage 1	Stage 2	Stage 3
Public administrations	19.439	19.163	200	76	(48)	(14)	(9)	(25)	19.391	19.149	191	51
Other financial corporations	9.856	9.747	95	14	(39)	(25)	(6)	(7)	9.817	9.722	88	7
Non-financial corporations	142.547	119.891	15.179	7.477	(6.123)	(774)	(1.110)	(4.239)	136.424	119.117	14.069	3.238
Individuals	151.410	129.196	15.108	7.106	(5.895)	(1.192)	(1.161)	(3.542)	145.515	128.005	13.946	3.564
Loans and advances to customers	323.252	277.998	30.581	14.672	(12.105)	(2.005)	(2.287)	(7.813)	311.147	275.993	28.294	6.860

December 2019 (Millions of Euros)
	Gross exposure				Accumulated allowances				Net amount
	Total	Stage 1	Stage 2	Stage 3	Total	Stage 1	Stage 2	Stage 3	Total	Stage 1	Stage 2	Stage 3
Public administrations	28.281	27.511	682	88	(59)	(15)	(22)	(21)	28.222	27.496	660	66
Other financial corporations	11.239	11.085	136	17	(31)	(19)	(2)	(10)	11.207	11.066	134	8
Non-financial corporations	173.254	148.768	16.018	8.468	(6.465)	(811)	(904)	(4.750)	166.789	147.957	15.114	3.718
Individuals	181.989	158.085	16.523	7.381	(5.847)	(1.283)	(1.252)	(3.312)	176.142	156.801	15.272	4.069
Loans and advances to customers	394.763	345.449	33.360	15.954	(12.402)	(2.129)	(2.181)	(8.093)	382.360	343.320	31.179	7.861

December 2018 (Millions of Euros)
	Gross exposure				Accumulated allowances				Net amount
	Total	Stage 1	Stage 2	Stage 3	Total	Stage 1	Stage 2	Stage 3	Total	Stage 1	Stage 2	Stage 3
Public administrations	28.632	27.740	764	128	(84)	(21)	(25)	(38)	28.549	27.719	739	91
Other financial corporations	9.490	9.189	291	11	(22)	(13)	(4)	(4)	9.468	9.176	286	6
Non-financial corporations	169.764	145.875	15.516	8.372	(6.260)	(730)	(1.190)	(4.341)	163.503	145.145	14.327	4.031
Individuals	178.339	156.400	14.102	7.838	(5.833)	(1.305)	(1.155)	(3.372)	172.506	155.094	12.946	4.466
Loans and advances to customers	386.225	339.204	30.673	16.348	(12.199)	(2.070)	(2.374)	(7.755)	374.027	337.134	28.299	8.593

The breakdown by counterparty and product of loans and advances, net of loss allowances, as well as the gross carrying amount by type of product, classified in different headings of the assets, as of December 31, 2020, 2019 and 2018 is shown below:

December 2020 (Millions of Euros)
	Central banks	General governments	Credit institutions	Other financial corporations	Non-financial corporations	Households	Total	Gross carrying amount
On demand and short notice	-	7	-	502	1.798	528	2.835	3.021
Credit card debt	-	-	-	2	1.485	11.605	13.093	14.220
Commercial debtors		898	-	317	14.262	67	15.544	15.796
Finance leases	-	197	-	6	7.125	322	7.650	8.013
Reverse repurchase loans	472	-	1.914	-	71	-	2.457	2.463
Other term loans	5.690	18.111	3.972	5.799	111.141	132.603	277.317	287.467
Advances that are not loans	48	260	8.721	3.191	1.084	473	13.777	13.833
LOANS AND ADVANCES	6.209	19.475	14.608	9.817	136.966	145.598	332.672	344.813
By secured loans
Of which: mortgage loans collateralized by immovable property		372	-	209	22.091	94.147	116.819	120.194
Of which: other collateralized loans	472	952	-	317	3.763	2.059	7.562	7.776
By purpose of the loan
Of which: credit for consumption						39.799	39.799	43.037
Of which: lending for house purchase						94.098	94.098	95.751
By subordination
Of which: project finance loans					10.721		10.721	11.032

December 2019 (Millions of Euros)
	Central banks	General governments	Credit institutions	Other financial corporations	Non-financial corporations	Households	Total	Gross carrying amount
On demand and short notice	-	9	-	118	2.328	595	3.050	3.251
Credit card debt	-	10	1	3	1.940	14.401	16.355	17.608
Commercial debtors		971	-	230	15.976	99	17.276	17.617
Finance leases	-	227	-	6	8.091	387	8.711	9.095
Reverse repurchase loans	-	-	1.817	-	26	-	1.843	1.848
Other term loans	4.240	26.734	4.121	7.795	137.934	160.223	341.047	351.230
Advances that are not loans	35	865	7.743	3.056	951	506	13.156	13.214
LOANS AND ADVANCES	4.275	28.816	13.682	11.208	167.246	176.211	401.438	413.863
By secured loans
Of which: mortgage loans collateralized by immovable property		1.067	15	261	23.575	111.085	136.003	139.317
Of which: other collateralized loans	-	10.447	93	2.106	29.009	6.893	48.548	49.266
By purpose of the loan
Of which: credit for consumption						46.356	46.356	49.474
Of which: lending for house purchase						110.178	110.178	111.636
By subordination
Of which: project finance loans					12.259		12.259	12.415

December 2018 (Millions of Euros)
	Central banks	General governments	Credit institutions	Other financial corporations	Non-financial corporations	Households	Total	Gross carrying amount
On demand and short notice	-	10	-	151	2.833	648	3.641	3.834
Credit card debt	-	8	1	2	2.328	13.108	15.446	16.495
Commercial debtors		948	-	195	16.190	103	17.436	17.716
Finance leases	-	226	-	3	8.014	406	8.650	9.077
Reverse repurchase loans	-	293	477	-	-	-	770	772
Other term loans	3.911	26.839	2.947	7.030	133.573	157.760	332.060	342.264
Advances that are not loans	29	1.592	5.771	2.088	984	498	10.962	11.025
LOANS AND ADVANCES	3.941	29.917	9.196	9.468	163.922	172.522	388.966	401.183
By secured loans
Of which: mortgage loans collateralized by immovable property		1.056	15	219	26.784	111.809	139.883	144.005
Of which: other collateralized loans	-	7.179	285	1.389	31.393	6.835	47.081	47.855
By purpose of the loan
Of which: credit for consumption						40.124	40.124	42.736
Of which: lending for house purchase						111.007	111.007	112.952
By subordination
Of which: project finance loans					13.973		13.973	14.286

7.2.3. Mitigation of credit risk, collateralized credit risk and other credit enhancements

In most cases, maximum credit risk exposure is reduced by collateral, credit enhancements and other actions which mitigate the Group’s exposure. The BBVA Group applies a credit risk hedging and mitigation policy deriving from a banking approach focused on relationship banking. The existence of guarantees could be a necessary but not sufficient instrument for accepting risks, as the assumption of risks by the Group requires prior evaluation of the debtor’s capacity for repayment, or that the debtor can generate sufficient resources to allow the amortization of the risk incurred under the agreed terms.

The policy of accepting risks is therefore organized into three different levels in the BBVA Group:

Analysis of the financial risk of the transaction, based on the debtor’s capacity for repayment or generation of funds.

The constitution of guarantees that are adequate, or at any rate generally accepted, for the risk assumed, in any of the generally accepted forms: monetary, secured, personal or hedge guarantees; and finally

Assessment of the repayment risk (asset liquidity) of the guarantees received.

This is carried out through a prudent risk policy that consists of the analysis of the financial risk, based on the capacity for reimbursement or generation of resources of the borrower, the analysis of the guarantee, assessing, among others, the efficiency, the robustness and the risk, the adequacy of the guarantee with the operation and other aspects such as the location, currency, concentration or the existence of limitations. Additionally, the necessary tasks for the constitution of guarantees must be carried out - in any of the generally accepted forms (collaterals, personal guarantees and financial hedge instruments) - appropriate to the risk assumed.

The procedures for the management and valuation of collateral are set out in the corporate general policies (retail and wholesale), which establish the basic principles for credit risk management, including the management of collaterals assigned in transactions with customers. The criteria for the systematic, standardized and effective treatment of collateral in credit transaction procedures in BBVA Group’s wholesale and retail banking are included in the Specific Collateral Rules.

The methods used to value the collateral are in line with the best market practices and imply the use of appraisal of real-estate collateral, the market price in market securities, the trading price of shares in mutual funds, etc. All the collaterals received must be correctly assigned and entered in the corresponding register. They must also have the approval of the Group’s legal units.

The valuation of the collateral is taken into account in the calculation of the expected losses. The Group has developed internal models to estimate the realization value of the collaterals received, the time that elapses until then, the costs for their acquisition, maintenance and subsequent sale, from real observations based on its own experience. This modeling is part of the LGD estimation processes that are applied to the different segments, and is included within the annual review and validation procedures.

The following is a description of the main types of collateral for each financial instrument class:

Debt instruments held for trading: The guarantees or credit enhancements obtained directly from the issuer or counterparty are implicit in the clauses of the instrument(mainly guarantees of the issuer).

Derivatives and hedging derivatives: In derivatives, credit risk is minimized through contractual netting agreements, where positive- and negative-value derivatives with the same counterparty are offset for their net balance. There may likewise be other kinds of guarantees and collaterals, depending on counterparty solvency and the nature of the transaction (mainly collaterals).

The summary of the offsetting effect (via netting and collateral) for derivatives and securities operations as of December 31, 2020 is presented in Note 7.3.2.

Other financial assets designated at fair value through profit or loss and financial assets at fair value through other comprehensive income: The guarantees or credit enhancements obtained directly from the issuer or counterparty are inherent to the structure of the instrument (mainly personal guarantees).

As of December 31, 2020, 2019 and 2018, BBVA Group had no credit risk exposure of impaired financial assets at fair value through other comprehensive income (see Note 7.2.2).

Financial assets at amortized cost:

Loans and advances to credit institutions: These usually have the counterparty’s personal guarantee or pledged securities in the case of repos.

Loans and advances to customers: Most of these loans and advances are backed by personal guarantees extended by the customer. There may also be collateral to secure loans and advances to customers (such as mortgages, cash collaterals, pledged securities and other collateral), or to obtain other credit enhancements (bonds or insurances).

Debt securities: The guarantees or credit enhancements obtained directly from the issuer or counterparty are inherent to the structure of the instrument.

Financial guarantees, other contingent risks and drawable by third parties: these have the counterparty’s personal guarantee or other types of collaterals.

The disclosure of impaired loans and advances at amortized cost covered by collateral (see Note 7.2.6), by type of collateral, as of December 31, 2020, 2019 and 2018, is the following:

December 2020 (Millions of Euros)
	Maximum exposure to credit risk	Of which secured by collateral
		Residential properties	Commercial properties	Cash	Others	Financial
Impaired loans and advances at amortized cost	14.678	2.717	789	18	52	575
Total	14.678	2.717	789	18	52	575

December 2019 (Millions of Euros)
	Maximum exposure to credit risk	Of which secured by collateral
		Residential properties	Commercial properties	Cash	Others	Financial
Impaired loans and advances at amortized cost	15.959	3.396	939	35	221	542
Total	15.959	3.396	939	35	221	542

December 2018 (Millions of Euros)
	Maximum exposure to credit risk	Of which secured by collateral
		Residential properties	Commercial properties	Cash	Others	Financial
Impaired loans and advances at amortized cost	16.359	3.484	1.255	13	317	502
Total	16.359	3.484	1.255	13	317	502

The value of guarantees received as of December 31, 2020, 2019 and 2018, is the following:

Guarantees received (Millions of Euros)
	2020	2019	2018
Value of collateral	116.900	152.454	158.268
Of which: guarantees normal risks under special monitoring	11.296	14.623	14.087
Of which: guarantees non-performing risks	3.577	4.590	5.068
Value of other guarantees	47.012	35.464	16.897
Of which: guarantees normal risks under special monitoring	4.045	3.306	1.519
Of which: guarantees non-performing risks	575	542	502
Total value of guarantees received	163.912	187.918	175.165

The maximum credit risk exposure of impaired financial guarantees and other commitments at December 31, 2020, 2019 and 2018 amounts to €1,032, €1,001 and €987 million, respectively (see Note 7.2.2).

7.2.4. Credit quality of financial assets that are neither past due nor impaired

The BBVA Group has tools that enable it to rank the credit quality of its transactions and customers based on an assessment and its correspondence with the probability of default (“PD”) scales. To analyze the performance of PD, the Group has a series of tracking tools and historical databases that collect the pertinent internally generated information. These tools can be grouped together into scoring and rating models.

Scoring

Scoring is a decision-making model that contributes to both the arrangement and management of retail loans: consumer loans, mortgages, credit cards for individuals, etc. Scoring is the tool used to decide to originate a loan, what amount should be originated and what strategies can help establish the price, because it is an algorithm that sorts transactions by their credit quality. This algorithm enables the BBVA Group to assign a score to each transaction requested by a customer, on the basis of a series of objective characteristics that have statistically been shown to discriminate between the quality and risk of this type of transactions. The advantage of scoring lies in its simplicity and homogeneity: all that is needed is a series of objective data for each customer, and this data is analyzed automatically using an algorithm.

There are three types of scoring, based on the information used and on its purpose:

Reactive scoring: measures the risk of a transaction requested by an individual using variables relating to the requested transaction and to the customer’s socio-economic data available at the time of the request. The new transaction is approved or rejected depending on the score.

Behavioral scoring: scores transactions for a given product in an outstanding risk portfolio of the entity, enabling the credit rating to be tracked and the customer’s needs to be anticipated. It uses transaction and customer variables available internally. Specifically, variables that refer to the behavior of both the product and the customer.

Proactive scoring: gives a score at customer level using variables related to the individual’s general behavior with the entity, and to his/her payment behavior in all the contracted products. The purpose is to track the customer’s credit quality and it is used to pre-approve new transactions.

Rating

Rating tools, as opposed to scoring tools, do not assess transactions but focus on the rating of customers instead: companies, corporations, SMEs, general governments, etc. A rating tool is an instrument that, based on a detailed financial study, helps determine a customer’s ability to meet his/her financial obligations. The final rating is usually a combination of various factors: on one hand, quantitative factors, and on the other hand, qualitative factors. It is a middle road between an individual analysis and a statistical analysis.

The main difference between ratings and scorings is that the latter are used to assess retail products, while ratings use a wholesale banking customer approach. Moreover, scorings only include objective variables, while ratings add qualitative information. And although both are based on statistical studies, adding a business view, rating tools give more weight to the business criterion compared to scoring tools.

For portfolios where the number of defaults is low (sovereign risk, corporates, financial entities, etc.) the internal information is supplemented by “benchmarking” of the external rating agencies (Moody’s, Standard & Poor’s and Fitch). To this end, each year the PDs compiled by the rating agencies at each level of risk rating are compared, and the measurements compiled by the various agencies are mapped against those of the BBVA master rating scale.

Once the probability of default of a transaction or customer has been calculated, a "business cycle adjustment" is carried out. This is a means of establishing a measure of risk that goes beyond the time of its calculation. The aim is to capture representative information of the behavior of portfolios over a complete economic cycle. This probability is linked to the Master Rating Scale prepared by the BBVA Group to enable uniform classification of the Group’s various asset risk portfolios.

The table below shows the abridged scale used to classify the BBVA Group’s outstanding risk as of December 31, 2020:

External rating	Internal rating	Probability of default(basis points)
Standard&Poor's List	Reduced List (22 groups)	Average	Minimum from >=	Maximum
AAA	AAA	1	-	2
AA+	AA+	2	2	3
AA	AA	3	3	4
AA-	AA-	4	4	5
A+	A+	5	5	6
A	A	8	6	9
A-	A-	10	9	11
BBB+	BBB+	14	11	17
BBB	BBB	20	17	24
BBB-	BBB-	31	24	39
BB+	BB+	51	39	67
BB	BB	88	67	116
BB-	BB-	150	116	194
B+	B+	255	194	335
B	B	441	335	581
B-	B-	785	581	1.061
CCC+	CCC+	1.191	1.061	1.336
CCC	CCC	1.500	1.336	1.684
CCC-	CCC-	1.890	1.684	2.121
CC+	CC+	2.381	2.121	2.673
CC	CC	3.000	2.673	3.367
CC-	CC-	3.780	3.367	4.243

These different levels and their probability of default were calculated by using as a reference the rating scales and default rates provided by the external agencies Standard & Poor’s and Moody’s. These calculations establish the levels of probability of default for the BBVA Group’s Master Rating Scale. Although this scale is common to the entire Group, the calibrations (mapping scores to PD sections/Master Rating Scale levels) are carried out at tool level for each country in which the Group has tools available.

The table below outlines the distribution by probability of default within 12 months and stages of the gross carrying amount of loans and advances to customers in percentage terms of the BBVA Group as of December 31, 2020, 2019 and 2018:

Probability of default (basis points)
	2020		2019		2018
	Subject to 12 month ECL (Stage 1)	Subject to lifetime ECL (Stage 2)	Subject to 12 month ECL (Stage 1)	Subject to lifetime ECL (Stage 2)	Subject to 12 month ECL (Stage 1)	Subject to lifetime ECL (Stage 2)
	%	%	%	%	%	%
0 to 2	4,0	-	5,5	-	9,6	-
2 to 5	10,2	0,1	6,3	-	10,8	0,1
5 to 11	7,7	0,1	14,6	0,2	6,3	-
11 to 39	26,8	0,5	24,5	0,8	20,9	0,4
39 to 194	24,0	2,3	24,5	1,6	30,1	1,8
194 to 1,061	15,1	3,4	14,0	3,6	12,2	3,6
1,061 to 2,121	1,5	1,2	1,4	1,2	1,6	1,2
> 2,121	0,6	2,5	0,4	1,5	0,2	1,2
Total	89,9	10,1	91,0	9,0	91,7	8,3

7.2.5. Impaired secured loan risks

The breakdown of loans and advances, within financial assets at amortized cost, non-performing and accumulated impairment, as well as the gross carrying amount, by counterparties as of December 31, 2020, 2019 and 2018 is as follows:

December 2020 (Millions of Euros)
	Gross carrying amount	Non-performing loans and advances	Accumulated impairment	Non-performing loans and advances as a % of the total
Central banks	6.229	-	(20)	-
General governments	19.439	76	(48)	0,4%
Credit institutions	14.591	6	(16)	-
Other financial corporations	9.856	14	(39)	0,1%
Non-financial corporations	142.547	7.477	(6.123)	5,2%
Agriculture, forestry and fishing	3.438	132	(108)	3,8%
Mining and quarrying	4.349	47	(59)	1,1%
Manufacturing	33.771	1.486	(1.129)	4,4%
Electricity, gas, steam and air conditioning supply	13.490	591	(509)	4,4%
Water supply	899	17	(15)	1,9%
Construction	10.019	1.397	(722)	13,9%
Wholesale and retail trade	24.594	1.456	(1.223)	5,9%
Transport and storage	8.117	489	(368)	6,0%
Accommodation and food service activities	8.337	358	(294)	4,3%
Information and communications	5.764	73	(60)	1,3%
Financial and insurance activities	5.298	123	(132)	2,3%
Real estate activities	10.025	617	(494)	6,2%
Professional, scientific and technical activities	2.886	177	(124)	6,1%
Administrative and support service activities	3.955	142	(192)	3,6%
Public administration and defense; compulsory social security	129	5	(4)	3,5%
Education	665	54	(43)	8,1%
Human health services and social work activities	1.812	67	(59)	3,7%
Arts, entertainment and recreation	1.131	46	(65)	4,1%
Other services	3.871	198	(523)	5,1%
Households	151.410	7.106	(5.895)	4,7%
LOANS AND ADVANCES	344.072	14.678	(12.141)	4,3%

December 2019 (Millions of Euros)
	Gross carrying amount	Non-performing loans and advances	Accumulated impairment	Non-performing loans and advances as a % of the total
Central banks	4.285	-	(9)	-
General governments	28.281	88	(60)	0,3%
Credit institutions	13.664	6	(15)	-
Other financial corporations	11.239	17	(31)	0,2%
Non-financial corporations	173.254	8.467	(6.465)	4,9%
Agriculture, forestry and fishing	3.758	154	(124)	4,1%
Mining and quarrying	4.669	100	(86)	2,1%
Manufacturing	39.517	1.711	(1.242)	4,3%
Electricity, gas, steam and air conditioning supply	12.305	684	(575)	5,6%
Water supply	900	14	(16)	1,6%
Construction	10.945	1.377	(876)	12,6%
Wholesale and retail trade	27.467	1.799	(1.448)	6,6%
Transport and storage	9.638	507	(392)	5,3%
Accommodation and food service activities	8.703	279	(203)	3,2%
Information and communications	6.316	95	(65)	1,5%
Financial and insurance activities	6.864	191	(140)	2,8%
Real estate activities	19.435	782	(527)	4,0%
Professional, scientific and technical activities	4.375	167	(140)	3,8%
Administrative and support service activities	3.415	118	(134)	3,4%
Public administration and defense, compulsory social security	282	5	(6)	1,7%
Education	903	41	(38)	4,5%
Human health services and social work activities	4.696	66	(55)	1,4%
Arts, entertainment and recreation	1.396	47	(39)	3,4%
Other services	7.671	331	(360)	4,3%
Households	181.989	7.381	(5.847)	4,1%
LOANS AND ADVANCES	412.711	15.959	(12.427)	3,9%

December 2018 (Millions of Euros)
	Gross carrying amount	Non-performing loans and advances	Accumulated impairment	Non-performing loans and advances as a % of the total
Central banks	3.947	-	(6)	-
General governments	28.198	128	(84)	0,4%
Credit institutions	9.175	10	(12)	0,1%
Other financial corporations	9.490	11	(22)	0,1%
Non-financial corporations	170.182	8.372	(6.260)	4,9%
Agriculture, forestry and fishing	3.685	122	(107)	3,3%
Mining and quarrying	4.952	96	(70)	1,9%
Manufacturing	36.772	1.695	(1.134)	4,6%
Electricity, gas, steam and air conditioning supply	13.853	585	(446)	4,2%
Water supply	1.061	19	(15)	1,8%
Construction	11.899	1.488	(1.007)	12,5%
Wholesale and retail trade	25.833	1.624	(1.259)	6,3%
Transport and storage	9.798	459	(374)	4,7%
Accommodation and food service activities	7.882	315	(204)	4,0%
Information and communications	5.238	113	(72)	2,1%
Financial and insurance activities	6.929	147	(128)	2,1%
Real estate activities	17.272	834	(624)	4,8%
Professional, scientific and technical activities	5.096	204	(171)	4,0%
Administrative and support service activities	3.162	128	(125)	4,0%
Public administration and defense, compulsory social security	319	5	(7)	1,6%
Education	912	31	(31)	3,4%
Human health services and social work activities	4.406	63	(63)	1,4%
Arts, entertainment and recreation	1.323	59	(41)	4,5%
Other services	9.791	386	(382)	3,9%
Households	178.355	7.838	(5.833)	4,4%
LOANS AND ADVANCES	399.347	16.359	(12.217)	4,1%

The changes during the years 2020, 2019 and 2018 of impaired financial assets and contingent risks are as follow:

Changes in impaired financial assets and contingent risks (Millions of Euros)
	2020	2019	2018
Balance at the beginning	16.770	17.134	20.590
Additions	9.533	9.857	9.792
Decreases (*)	(5.024)	(5.874)	(6.909)
Net additions	4.509	3.983	2.883
Amounts written-off	(3.603)	(3.803)	(5.076)
Exchange differences and other	(968)	(544)	(1.264)
Balance at the end	16.708	16.770	17.134

(*) Reflects the total amount of impaired loans derecognized from the consolidated balance sheet throughout the year as a result of mortgage foreclosures and real estate assets received in lieu of payment as well as monetary recoveries.

The changes during the years 2020, 2019 and 2018 in financial assets derecognized from the accompanying consolidated balance sheet as their recovery is considered unlikely ("write-offs"), is shown below:

Changes in impaired financial assets written-off from the balance sheet (Millions of Euros)
	Notes	2020	2019	2018
Balance at the beginning		26.245	32.343	30.139
Companies held for sale (*)		(4.646)	-	-
Increase		3.440	4.712	6.164
Decrease:		(2.715)	(11.039)	(4.210)
Re-financing or restructuring		(7)	(2)	(10)
Cash recovery	47	(339)	(919)	(589)
Foreclosed assets		(479)	(617)	(625)
Sales (**)		(1.223)	(8.325)	(1.805)
Debt forgiveness		(607)	(493)	(889)
Time-barred debt and other causes		(60)	(682)	(292)
Net exchange differences		(323)	230	250
Balance at the end		22.001	26.245	32.343

(*) Amount in 2020 is mainly due to the sale of the stake in BBVA USA (see Notes 3 and 21).

(**) Includes principal and interest.

As indicated in Note 2.2.1, although they have been derecognized from the consolidated balance sheet, the BBVA Group continues to attempt to collect on these written-off financial assets, until the rights to receive them are fully extinguished, either because it is a time-barred financial asset, the financial asset is forgiven, or other reason.

7.2.6. Loss allowances

Movements in gross accounting balances and accumulated allowances for loan losses during 2020 and 2019 are recorded on the accompanying consolidated balance sheet as of December 31, 2020 and 2019, in order to cover the estimated loss allowances in loans and advances and debt securities measured at amortized cost.

Changes in gross accounting balances of loans and advances at amortized cost. 2020 (Millions of Euros)
	Stage 1	Stage 2	Stage 3	Total
Opening balance	363.234	33.518	15.959	412.711
Transfers of financial assets:	(11.935)	8.807	3.128	-
Transfers from stage 1 to Stage 2	(15.843)	15.843	-	-
Transfers from stage 2 to Stage 1	5.107	(5.107)	-	-
Transfers to Stage 3	(1.701)	(2.659)	4.359	-
Transfers from Stage 3	502	729	(1.231)	-
Net annual origination of financial assets	16.119	(827)	102	15.395
Becoming write-offs	(3)	(2)	(2.944)	(2.949)
Foreign exchange	(21.472)	(2.342)	(1.157)	(24.970)
Modifications that do not result in derecognition	(204)	827	511	1.134
Other	(283)	(190)	270	(204)
Discontinued operations	(46.664)	(9.190)	(1.192)	(57.045)
Closing balance	298.793	30.601	14.678	344.072

Changes in allowances of loans and advances at amortized cost. 2020 (Millions of Euros)
	Stage 1	Stage 2	Stage 3	Total
Opening balance	(2.149)	(2.183)	(8.094)	(12.427)
Transfers of financial assets:	184	(511)	(1.806)	(2.133)
Transfers from stage 1 to Stage 2	156	(923)	-	(766)
Transfers from stage 2 to Stage 1	(50)	253	-	202
Transfers to Stage 3	81	218	(1.950)	(1.652)
Transfers from Stage 3	(3)	(59)	144	83
Net annual origination of allowances	(872)	(795)	(1.329)	(2.996)
Becoming write-offs	-	-	2.567	2.568
Foreign exchange	227	256	721	1.204
Modifications that do not result in derecognition	12	(118)	(177)	(283)
Other	160	618	25	803
Discontinued operations	401	444	278	1.123
Closing balance	(2.037)	(2.289)	(7.815)	(12.141)

Changes in gross accounting balances of loans and advances at amortized cost. 2019 (Millions of Euros)
	Stage 1	Stage 2	Stage 3	Total
Opening balance	352.282	30.707	16.359	399.347
Transfers of financial assets:	(9.021)	6.279	2.741	-
Transfers from stage 1 to Stage 2	(13.546)	13.546	-	-
Transfers from stage 2 to Stage 1	5.656	(5.656)	-	-
Transfers to Stage 3	(1.571)	(2.698)	4.269	-
Transfers from Stage 3	440	1.087	(1.527)	-
Net annual origination of financial assets	20.296	(2.739)	246	17.804
Becoming write-offs	(152)	(349)	(3.407)	(3.908)
Foreign exchange	1.611	35	16	1.662
Modifications that do not result in derecognition	(1)	(27)	15	(13)
Other	(1.782)	(388)	(11)	(2.180)
Closing balance	363.234	33.518	15.959	412.711

Changes in allowances of loans and advances at amortized cost. 2019 (Millions of Euros)
	Stage 1	Stage 2	Stage 3	Total
Opening balance	(2.082)	(2.375)	(7.761)	(12.217)
Transfers of financial assets:	176	(227)	(1.574)	(1.626)
Transfers from stage 1 to Stage 2	126	(649)	-	(523)
Transfers from stage 2 to Stage 1	(38)	273	-	235
Transfers to Stage 3	89	234	(1.810)	(1.487)
Transfers from Stage 3	(1)	(86)	236	149
Net annual origination of allowances	(542)	(116)	(1.711)	(2.370)
Becoming write-offs	130	337	2.789	3.256
Foreign exchange	(30)	(18)	69	20
Modifications that do not result in derecognition	(15)	(149)	(89)	(254)
Other	215	366	183	764
Closing balance	(2.149)	(2.183)	(8.094)	(12.427)

The following are the movements produced during 2018 in the value adjustments recorded in the accompanying balance sheets to cover the impairment or reversal of the estimated impairment of financial assets at amortized cost:

Financial assets at amortized cost. December 2018 (Millions of Euros)
	Not credit-impaired			Credit-impaired	Total
	Stage 1	Stage 2		Credit-impaired (Stage 3)
	Loss allowances	Loss allowances (collectively assessed)	Loss allowances (individually assessed)	Loss allowances	Loss allowances
Opening balance	(2.237)	(1.827)	(525)	(9.371)	(13.960)
Transfers of financial assets:	131	(155)	328	(1.794)	(1.490)
Transfers from Stage 1 to Stage 2 (not credit-impaired)	208	(930)	(218)	-	(940)
Transfers from Stage 2 (not credit - impaired) to Stage 1	(125)	619	50	-	544
Transfers to Stage 3	55	282	564	(2.127)	(1.226)
Transfers from Stage 3 to Stage 1 or 2	(7)	(126)	(68)	333	132
Changes without transfers between Stages	358	(53)	(260)	(3.775)	(3.730)
New financial assets originated	(1.072)	(375)	(244)	-	(1.692)
Disposals	2	3	-	110	115
Repayments	641	432	118	1.432	2.623
Write-offs	13	14	2	4.433	4.461
Foreign exchange	(84)	72	(93)	343	239
Modifications that result in derecognition	5	10	25	98	138
Modifications that do not result in derecognition	3	(8)	1	(362)	(366)
Other	135	133	20	1.111	1.399
Closing balance	(2.106)	(1.753)	(628)	(7.777)	(12.264)
Of which: Loans and advances					(12.217)
Of which: Debt certificates					(46)

7.2.7. Refinancing and restructuring transactions

Group policies and principles with respect to refinancing and restructuring transactions

Refinancing and restructuring transactions (see definition in the Glossary) are carried out with customers who have requested such a transaction in order to meet their current loan payments if they are expected, or may be expected, to experience financial difficulty in making the payments in the future.

The basic aim of a refinancing and restructuring transaction is to provide the customer with a situation of financial viability over time by adapting repayment of the loan incurred with the Group to the customer’s new situation of fund generation. The use of refinancing and restructuring for other purposes, such as to delay loss recognition, is contrary to BBVA Group policies.

The BBVA Group’s refinancing and restructuring policies are based on the following general principles:

Refinancing and restructuring is authorized according to the capacity of customers to pay the new installments. This is done by first identifying the origin of the payment difficulties and then carrying out an analysis of the customers’ viability, including an updated analysis of their economic and financial situation and capacity to pay and generate funds. If the customer is a company, the analysis also covers the situation of the industry in which it operates.

With the aim of increasing the solvency of the transaction, new guarantees and/or guarantors of demonstrable solvency are obtained where possible. An essential part of this process is an analysis of the effectiveness of both the new and original guarantees.

This analysis is carried out from the overall customer or group perspective.

Refinancing and restructuring transactions do not in general increase the amount of the customer’s loan, except for the expense inherent to the transaction itself.

The capacity to refinance and restructure a loan is not delegated to the branches, but decided on by the risk units.

The decisions made are reviewed from time to time with the aim of evaluating full compliance with refinancing and restructuring policies.

These general principles are adapted in each case according to the conditions and circumstances of each geographical area in which the Group operates, and to the different types of customers involved.

In the case of retail customers (private individuals), the main aim of the BBVA Group’s policy on refinancing and restructuring a loan is to avoid default arising from a customer’s temporary liquidity problems by implementing structural solutions that do not increase the balance of the customer’s loan. The solution required is adapted to each case and the loan repayment is made easier, in accordance with the following principles:

Analysis of the viability of transactions based on the customer’s willingness and ability to pay, which may be reduced, but should nevertheless be present. The customer must therefore repay at least the interest on the transaction in all cases. No arrangements may be concluded that involve a grace period for both principal and interest.

Refinancing and restructuring of transactions is only allowed on those loans in which the BBVA Group originally entered into.

Customers subject to refinancing and restructuring transactions are excluded from marketing campaigns of any kind.

In the case of non-retail customers (mainly companies, enterprises and corporates), refinancing/restructuring is authorized according to an economic and financial viability plan based on:

Forecasted future income, margins and cash flows to allow entities to implement cost adjustment measures (industrial restructuring) and a business development plan that can help reduce the level of leverage to sustainable levels (capacity to access the financial markets).

Where appropriate, the existence of a divestment plan for assets and/or operating segments that can generate cash to assist the deleveraging process.

The capacity of shareholders to contribute capital and/or guarantees that can support the viability of the plan.

In accordance with the Group’s policy, the conclusion of a loan refinancing and restructuring transaction does not mean the loan is reclassified from "impaired" or "significant increase in credit risk" to normal risk. The reclassification to "significant increase in credit risk" or normal risk categories must be based on the analysis mentioned earlier of the viability, upon completion of the probationary periods described below.

The Group maintains the policy of including risks related to refinanced and restructured loans as either:

"Impaired assets", as although the customer is up to date with payments, they are classified as unlikely to pay when there are significant doubts that the terms of their refinancing may not be met; or

"Significant increase in credit risk" until the conditions established for their consideration as normal risk are met.

The assets classified as "Impaired assets" should comply with the following conditions in order to be reclassified to "Significant increase in credit risk":

The customer has to have paid a significant part of the pending exposure.

At least one year must have elapsed since its classification as "Impaired asset".

The customer does not have past due payments and objective criteria, demonstrating the borrower´s ability to pay, have been verified.

The conditions established for assets classified as “Significant increase in credit risk” to be reclassified out of this category are as follows:

The customer must have paid past-due amounts (principal and interest) since the date of the renegotiation or restructuring of the loan or other objective criteria, demonstrating the borrower´s ability to pay, have been verified; none of its exposures is more than 30 days past-due.

At least two years must have elapsed since completion of the renegotiation or restructuring of the loan and regular payments must have been made during at least half of this probation period; and

It is unlikely that the customer will have financial difficulties and, therefore, it is expected that the customer will be able to meet its loan payment obligations (principal and interest) in a timely manner.

The economic impact caused by the Covid-19 pandemic has required the adaptation of the repayment schedule of a large volume of loans in all geographies and portfolios. In general, this support has been conducted through the concession of deferrals that comply with the principles established by the EBA, which has allowed for the application of a differential accounting and prudential treatment. Renewals and renegotiations will be classified as normal risk, provided that there is no significant increase in risk. This classification is applicable at the initial moment, and in the event of any deterioration, the criteria established in the existing governance are followed. In this sense, the aforementioned conditions are considered, including, among others, no facility with more than 30 days delinquency and not being identified as 'unlikely to pay'.

The BBVA Group’s refinancing and restructuring policy provides for the possibility of two modifications in a 24 month period for loans that are not in compliance with the payment schedule.

The internal models used to determine allowances for loan losses consider the restructuring and renegotiation of a loan, as well as re-defaults on such a loan, by assigning a lower internal rating to restructured and renegotiated loans than the average internal rating assigned to non-restructured/renegotiated loans. This downgrade results in an increase in the probability of default (PD) assigned to restructured/renegotiated loans (with the resulting PD being higher than the average PD of the non- renegotiated loans in the same portfolios).

For quantitative information on refinancing and restructuring transactions see Appendix VIII.

7.2.8. Risk concentration

Policies for preventing excessive risk concentration

In order to prevent the build-up of excessive risk concentrations at the individual, sector, portfolio and geography levels, BBVA Group maintains updated maximum permitted risk concentration indices which are tied to the various observable variables related to concentration risk.

Together with the limits for individual concentration, the Group uses the Herfindahl index to measure the concentration of the Group's portfolio and the banking group's subsidiaries. At the BBVA Group level, the index reached implies a "very low" degree of concentration.

The limit on the Group’s exposure or financial commitment to a specific customer therefore depends on the customer’s credit rating, the nature of the risks involved, and the Group’s presence in a given market, based on the following guidelines:

The aim is, as much as possible, to reconcile the customer's credit needs (commercial/financial, short-term/long-term, etc.) with the interests of the Group.

Any legal limits that may exist concerning risk concentration are taken into account (relationship between risks with a customer and the capital of the shareholder´s entity that assumes them), the markets, the macroeconomic situation, etc.

Risk concentrations by geography

The breakdown of the main figures in the most significant foreign currencies in the accompanying consolidated balance sheets is set forth in Appendix IX.

Sovereign risk concentration

Sovereign risk management

The risk associated with the transactions involving sovereign risk is identified, measured, controlled and tracked by a centralized unit integrated in the BBVA Group’s Risk Area. Its basic functions involve the preparation of reports in the countries where sovereign risk exists (called “financial programs”), tracking such risks, assigning ratings to these countries and, in general, supporting the Group in terms of reporting requirements for any transactions involving sovereign risk. The risk policies established in the financial programs are approved by the relevant risk committees.

The country risk unit tracks the evolution of the risks associated with the various countries to which the Group are exposed (including sovereign risk) on an ongoing basis in order to adapt its risk and mitigation policies to any macroeconomic and political changes that may occur. Moreover, it regularly updates its internal ratings and forecasts for these countries. The methodology is based on the assessment of quantitative and qualitative parameters which are in line with those used by certain multilateral organizations such as the International Monetary Fund (IMF) and the World Bank, rating agencies and export credit organizations.

For additional information on sovereign risk in Europe see Appendix IX.

Risk related to the developer and Real-Estate sector in Spain

The relative weight of the investment in Real Estate developments has dramatically decreased during the last years, especially since 2014 and during 2018, when doubtful assets exited the balance sheet and recovery of the sector concluded. A corporate sales policy has been rolled out to eliminate those real estate assets from the balance sheet which have been most difficult to commercialize. The sales of 80% of the Group’s share in Divarian and of other performing and NPL wholesale portfolios to Funds and specialized investors have been some of the most relevant transactions (see Note 3).

Policies and strategies established by the Group to deal with risks related to the developer and real-estate sector

BBVA Group has teams specializing in the management of the Real Estate Sector risk, given its economic importance and specific technical component. This specialization is not only in risk teams, but throughout the handling, commercial, problem risks and legal, etc. It also includes the research department of the BBVA Group (BBVA Research), which helps determine the medium/long-term vision needed to manage this portfolio.

The policies established to address the risks related to the developer and real-estate sector, aim to accomplish, among others, the following objectives: to avoid concentration in terms of customers, products and regions; to estimate the risk profile for the portfolio; and to anticipate possible worsening of the portfolio within a sector is highly cyclical.

Specific policies for analysis and granting of new developer risk transactions

In the analysis of new transactions, the assessment of the commercial operation in terms of the economic and financial viability of the project has been one of the constant.The monitoring of the work, sales prospects and the legal situation of the project are essential aspects for the admission and follow-up of new real estate transactions. With regard the participation of the Risk Acceptance teams, they have a direct link and participate in the committees of areas such as Valuation, Legal, Research and Recoveries. This guarantees coordination and exchange of information in all the processes.

In this context, and within the current Real Estate cycle, the strategy with clients is subject to an Asset Allocation limit and to an action framework that allows defining a target portfolio, both in volume and in credit quality.

Risk monitoring policies

The base information for analyzing the real estate portfolios is updated monthly. There is a systematic monitoring of developments under close monitoring with the evolution of works and sales. Since 2013, there are no threats of new defaults in the portfolio.

Policies applied in the management of real estate assets in Spain

The internal Rules on Real Estate Financing, which establish recommendations for financing a new housing development business, are reviewed and updated annually.

The recommendations represent guidelines about how to manage the credit admission activity of BBVA Group entities based on best practices of markets in which this activity is performed. It is expected that a high percentage of the current transactions will be in compliance with the latter.

For quantitative information about the risk related to the developer and Real-Estate sector in Spain see Appendix IX.

7.3. Market risk

Market risk originates from the possibility of experiencing losses in the value of positions held as a result of movements in market variables that affect the valuation of financial assets and liabilities. Market risk in the Group's trading portfolios stems mainly from the portfolios originated by Global Markets valued at fair value and held for the purpose of trading and generating short-term results. Market risk in the field of banking book is clearly and distinctly addressed and can be broken down into structural risks relating to interest rate, exchange rate and equity (see Note 7.4).

7.3.1. Market risk in trading portfolios

The main risks in the trading portfolios can be classified as follows:

Interest-rate risk: This arises as a result of exposure to movements in the different interest-rate curves involved in trading. Although the typical products that generate sensitivity to the movements in interest rates are money-market products (deposits, interest-rate futures, call money swaps, etc.) and traditional interest-rate derivatives (swaps and interest-rate options such as caps, floors, swaptions, etc.), practically all the financial products are exposed to interest-rate movements due to the effect that such movements have on the valuation of the financial discount.

Equity risk: This arises as a result of movements in share prices. This risk is generated in spot positions in shares or any derivative products whose underlying asset is a share or an equity index. Dividend risk is a sub-risk of equity risk, arising as an input for any equity option. Its variation may affect the valuation of positions and it is therefore a factor that generates risk on the books.

Exchange-rate risk: This is caused by movements in the exchange rates of the different currencies in which a position is held. As in the case of equity risk, this risk is generated in spot currency positions, and in any derivative product whose underlying asset is an exchange rate. In addition, the quanto effect (operations where the underlying asset and the instrument itself are denominated in different currencies) means that in certain transactions in which the underlying asset is not a currency, an exchange-rate risk is generated that has to be measured and monitored.

Credit-spread risk: Credit spread is an indicator of an issuer's credit quality. Spread risk occurs due to variations in the levels of spread of both corporate and government issues, and affects positions in bonds and credit derivatives.

Volatility risk: This occurs as a result of changes in the levels of implied price volatility of the different market instruments on which derivatives are traded. This risk, unlike the others, is exclusively a component of trading in derivatives and is defined as a first-order convexity risk that is generated in all possible underlying assets in which there are products with options that require a volatility input for their valuation.

The metrics developed to assess market risk in the BBVA Group are aligned with market practices and are implemented consistently across all the local market risk units.

Measurement procedures are established in terms of the possible impact of negative market conditions on the trading portfolio of the Group's Global Markets units, both under ordinary circumstances and in situations of heightened risk factors.

The standard metric used to measure market risk is Value at Risk (“VaR”), which indicates the maximum loss that may occur in the portfolios at a given confidence level (99%) and time horizon (one day). This statistic value is widely used in the market and has the advantage of summing up in a single metric the risks inherent to trading activity, taking into account how they are related and providing a prediction of the loss that the trading book could sustain as a result of fluctuations in equity prices, interest rates, foreign exchange rates and credit spreads. The market risk analysis considers various risks, such as credit spread risk, basis risk, as well as volatility and correlation risk.

With respect to the risk measurement models used by the BBVA Group, the Bank of Spain has authorized the use of the internal market risk model to determine bank capital requirements deriving from risk positions on theBBVA S.A. and BBVA Mexico trading book, which jointly accounted for around 72%, 72% and 76% of the Group’s trading-book market risk as of December 31, 2020, 2019 and 2018. For the rest of the geographical areas where the Group operates (applicable mainly to the Group´s South America subsidiaries, Garanti BBVA and BBVA USA), bank capital for the risk positions in the trading book is calculated using the Standardized Approach defined by the Basel Committee on Banking Supervision (which is referred to herein as the "standard model”).

The current management structure includes the monitoring of market-risk limits, consisting of a scheme of limits based on VaR, economic capital (based on VaR measurements) and VaR sub-limits, as well as stop-loss limits for each of the Group’s business units.

The model used estimates VaR in accordance with the historical simulation methodology, which involves estimating losses and gains that would have taken place in the current portfolio if the changes in market conditions that took place over a specific period of time in the past were repeated. Based on this information, it predicts the maximum expected loss of the current portfolio within a given confidence level. This model has the advantage of reflecting precisely the historical distribution of the market variables and not assuming any specific distribution of probability. The historical period used in this model is two years.

VaR figures are estimated with the following methodologies:

VaR without smoothing, which awards equal weight to the daily information for the previous two years. This is currently the official methodology for measuring market risks for the purpose of monitoring compliance with risk limits.

VaR with smoothing, which gives a greater weight to more recent market information. This metric supplements the previous one.

The use of VaR by historical simulation methodology as a risk metric has many advantages, but also certain limitations, among which it is worth highlighting:

The estimate of the maximum daily loss of the Global Markets portfolio positions (with a confidence level of 99%) depends on the market movements of the last two years, not picking up the impact of large market events if they have not occurred within that historical window

The use of the 99% confidence level does not consider potential losses that can occur beyond this level. To mitigate this limitation, different stress exercises are also performed, as described later.

At the same time, and following the guidelines established by the Spanish and European authorities, BBVA incorporates metrics in addition to VaR with the aim of meeting the Bank of Spain's regulatory requirements with respect to the calculation of bank capital for the trading book. Specifically, the measures incorporated in the Group since December 2011 (stipulated by Basel 2.5) are:

VaR: In regulatory terms, the VaR charge incorporates the stressed VaR charge, and the sum of the two (VaR and stressed VaR) is calculated. This quantifies the losses associated with the movements of the risk factors inherent to market operations (including interest-rate risk, exchange-rate risk, equity risk and credit risk, among others). Both VaR and stressed VaR are rescaled by a regulatory multiplier set at three and by the square root of ten to calculate the capital charge.

Specific Risk - Incremental Risk Capital (“IRC”) Quantification of the risks of default and downgrading of the credit ratings of the bond and credit derivative positions in the portfolio. The IRC charge is exclusively applied in entities in respect of which the internal market risk model is used (i.e. BBVA, S.A. and BBVA Mexico). The IRC charge is determined based on the associated losses (calculated at 99.9% confidence level over a one year horizon under the hypothesis of constant risk) due to a rating change and/or default of the issuer with respect to an asset. In addition, the price risk is included in sovereign positions for the specified items.

Specific Risk - Securitization and correlation portfolios. Capital charges for securitizations and correlation portfolios are assessed based on the potential losses associated with the rating level of a specific credit structure. They are calculated by the standard model. The scope of the correlation portfolios refers to the First To Default (FTD)-type market operation and/or tranches of market CDOs and only for positions with an active market and hedging capacity.

Validity tests are performed regularly on the risk measurement models used by the Group. They estimate the maximum loss that could have been incurred in the assessed positions with a certain level of probability (backtesting), as well as measurements of the impact of extreme market events on risk positions (stress testing). As an additional control measure, backtesting is conducted at a trading desk level in order to enable more specific monitoring of the validity of the measurement models.

Market risk in 2020

The Group’s market risk related to its trading portfolio remained at low levels compared to other risks managed by BBVA, particularly credit risk. This is due to the nature of the business. In 2020 the average VaR was €27 million, above the figure of 2019, with a maximum level in the year reached on the day May14, 2020 of €39 million. The evolution in the BBVA Group’s market risk during 2020, measured as VaR without smoothing (see Glossary) with a 99% confidence level and a 1-day horizon (shown in millions of Euros) is as follows:

By type of market risk assumed by the Group's trading portfolio, the main risk factor for the Group continued to be that linked to interest rates, with a weight of 56% of the total at December 31, 2020 (this figure includes the spread risk). The relative weight of this risk has increased compared with the close of 2019 (58%). Exchange-rate risk accounted for 22% of the total risk, increasing its weight with respect to December 2019 (13%), while equity, volatility and correlation risk has decreased, with a weight of 22% at the close of 2020 (vs. 29% at the close of 2019).

As of December 31, 2020, 2019 and 2018 the VaR was €32 million, €20 million and €17 million, respectively. The total VaR figures for 2020, 2019 and 2018 can be broken down as follows:

VaR by Risk Factor (Millions of Euros)
	Interest/Spread risk	Currency risk	Stock-market risk	Vega/Correlation risk	Diversification effect(*)	Total
2020
VaR average in the year	29	12	4	11	(28)	27
VaR max in the year	39	20	10	20	(14)	39
VaR min in the year	20	3	1	6	(39)	18
End of period VaR	32	12	2	11	(29)	28
2019
VaR average in the year	21	6	4	9	(20)	19
VaR max in the year	28	6	3	9	(21)	25
VaR min in the year	13	5	5	9	(18)	14
End of period VaR	24	5	5	8	(22)	20
2018
VaR average in the year	20	6	4	9	(20)	21
VaR max in the year	23	7	6	11	(21)	26
VaR min in the year	17	6	4	7	(18)	16
End of period VaR	19	5	3	7	(17)	17

(*) The diversification effect is the difference between the sum of the average individual risk factors and the total VaR figure that includes the implied correlation between all the variables and scenarios used in the measurement.

Validation of the internal market risk model

The internal market risk model is validated on a regular basis by backtesting in both, BBVA, S.A. and Global Markets Mexico (in BBVA Mexico). The aim of backtesting is to validate the quality and precision of the internal market risk model used by BBVA Group to estimate the maximum daily loss of a portfolio, at a 99% level of confidence and a 250-day time horizon, by comparing the Group's results and the risk measurements generated by the internal market risk model. These tests showed that the internal market risk model of both, BBVA, S.A. and Global Markets Mexico is adequate and precise.

Two types of backtesting have been carried out in 2020, 2019 and 2018:

"Hypothetical" backtesting: the daily VaR is compared with the results obtained, not taking into account the intraday results or the changes in the portfolio positions. This validates the appropriateness of the market risk metrics for the end-of-day position.

"Real" backtesting: the daily VaR is compared with the total results, including intraday transactions, but discounting the possible minimum charges or fees involved. This type of backtesting includes the intraday risk in portfolios.

In addition, each of these two types of backtesting was carried out at a risk factor or business type level, thus making a deeper comparison of the results with respect to risk measurements.

For the period between the year ended December 31, 2019 and the year ended December 31, 2020, the backtesting of the internal VaR calculation model was carried out, comparing the daily results obtained to the risk level estimated by the internal VaR calculation model. In that period, there were no negative exceptions in BBVA S.A., while in BBVA Mexico there were a total of 3 exceptions. The COVID-19 epidemic together with the fall in the oil price resulted in a sharp depreciation of the local currency, a considerable spike in stock market volatility, a breakdown of the correlation between different curves and an abrupt movement in local interest rate curves.

At the end of the year the comparison showed the internal VaR calculation model was working correctly, within the "green" zone (0-4 exceptions), thus validating the internal VaR calculation model, as has been the case each year since the internal market risk model was approved for the Group.

Stress testing

A number of stress tests are carried out on the BBVA Group's trading portfolios. First, global and local historical scenarios are used that replicate the behavior of an extreme past event, such as for example the collapse of Lehman Brothers or the "Tequilazo" crisis. These stress tests are complemented with simulated scenarios, where the aim is to generate scenarios that have a significant impact on the different portfolios, but without being anchored to any specific historical scenario. Finally, for some portfolios or positions, fixed stress tests are also carried out that have a significant impact on the market variables affecting these positions.

Historical scenarios

The historical benchmark stress scenario for the BBVA Group is Lehman Brothers, whose sudden collapse in September 2008 led to a significant impact on the behavior of financial markets at a global level. The following are the most relevant effects of this historical scenario:

Credit shock: reflected mainly in the increase of credit spreads and downgrades in credit ratings. 

Increased volatility in most of the financial markets (giving rise to a great deal of variation in the prices of different assets (currency, equity, debt).

Liquidity shock in the financial systems, reflected by a major movement in interbank curves, particularly in the shortest sections of the euro and dollar curves.

Simulated scenarios

Unlike the historical scenarios, which are fixed and therefore not suited to the composition of the risk portfolio at all times, the scenario used for the exercises of economic stress is based on resampling methodology. This methodology is based on the use of dynamic scenarios that are recalculated periodically depending on the main risks affecting the trading portfolios. On a data window wide enough to collect different periods of stress (data are taken from January 1, 2008 until the date of the assessment), a simulation is performed by resampling of historic observations, generating a distribution of losses and gains that serve to analyze the most extreme of births in the selected historical window. The advantage of this methodology is that the period of stress is not predetermined, but depends on the portfolio maintained at each time, and making a large number of simulations (10,000 simulations) allows a greater richness of information for the analysis of expected shortfall than what is available in the scenarios included in the calculation of VaR.

The main features of this approach are: a) the generated simulations respect the correlation structure of the data, b) there is flexibility in the inclusion of new risk factors and c) it allows the introduction of a lot of variability in the simulations (desirable for considering extreme events).

The impact of the stress test under multivariable simulation of the risk factors of the portfolio based on the expected shortfall (expected shortfall calculated at a 95% confidence level, 20 days) as of December 31, 2020 is as follows:

Impact of the stress test (Millions of Euros)
	Europe	Mexico	Peru	Venezuela	Argentina	Colombia	Turkey
Expected shortfall	(121)	(69)	(8)	-	(8)	(4)	(8)

7.3.2. Financial Instruments offset

Financial assets and liabilities may be netted in certain cases. In particular, they are presented for a net amount on the consolidated balance sheet only when the Group's entities satisfy the provisions of IAS 32-Paragraph 42, so they have both the legal right to net recognized amounts, and the intention of settling the net amount or of realizing the asset and simultaneously paying the liability.

In addition, the Group has presented as gross amounts assets and liabilities on the consolidated balance sheet for which there are master netting arrangements in place, but for which there is no intention of settling the net amount. The most common types of events that trigger the netting of reciprocal obligations are bankruptcy of the entity, surpassing certain level of indebtedness threshold, failure to pay, restructuring and dissolution of the entity.

In the current market context, derivatives are contracted under different framework contracts being the most widespread the ones developed by the International Swaps and Derivatives Association (“ISDA”) and, for the Spanish market, the Framework Agreement on Financial Transactions (“CMOF”). Almost all portfolio derivative transactions have been concluded under these framework contracts, including in them the netting clauses mentioned in the preceding paragraph as "Master Netting Agreement", greatly reducing the credit exposure on these instruments. Additionally, in contracts signed with counterparties, the collateral agreement annexes called Credit Support Annex (“CSA”) are included, thereby minimizing exposure to a potential default of the counterparty.

Moreover, many of the transactions involving assets purchased or sold under a repurchase agreement are transacted through clearing houses that articulate mechanisms to reduce counterparty risk, as well as through the signing of various master agreements for bilateral transactions, the most widely used being the Global Master Repurchase Agreement (GMRA), published by the International Capital Market Association (“ICMA”), to which the clauses related to the collateral exchange are usually added within the text of the master agreement itself.

A summary of the effect of offsetting (via netting and collateral) for derivatives and securities operations is presented below as of December 31, 2020, 2019 and 2018:

December 2020 (Millions of Euros)
					Gross amounts not offset in the consolidated balance sheets (D)
	Notes	Gross amounts recognized (A)	Gross amounts offset in the consolidated balance sheets (B)	Net amount presented in the consolidated balance sheets (C=A-B)	Financial instruments	Cash collateral received/ pledged	Net amount (E=C-D)
Trading and hedging derivatives	10, 15	47.862	5.688	42.173	33.842	9.018	(686)
Reverse repurchase, securities borrowing and similar agreements		34.500	-	34.500	35.141	161	(802)
Total assets		82.362	5.688	76.674	68.983	9.178	(1.488)
Trading and hedging derivatives	10, 15	49.720	5.722	43.998	33.842	9.435	721
Repurchase, securities lending and similar agreements		43.950	-	43.950	44.677	1.619	(2.346)
Total liabilities		93.670	5.722	87.948	78.519	11.054	(1.624)

December 2019 (Millions of Euros)
					Gross Amounts Not Offset in the Consolidated Balance Sheets (D)
	Notes	Gross amounts recognized (A)	Gross amounts offset in the consolidated balance sheets (B)	Net amount presented in the consolidated balance sheets (C=A-B)	Financial instruments	Cash collateral received/ pledged	Net amount (E=C-D)
Trading and hedging derivatives	10, 15	36.349	2.388	33.961	25.020	8.210	731
Reverse repurchase, securities borrowing and similar agreements		35.805	21	35.784	35.618	204	(39)
Total assets		72.154	2.409	69.744	60.637	8.415	692
Trading and hedging derivatives	10, 15	38.693	2.394	36.299	25.020	10.613	667
Repurchase, securities lending and similar agreements		45.977	21	45.956	45.239	420	297
Total liabilities		84.670	2.414	82.256	70.259	11.033	964

December 2019 (Millions of Euros)
					Gross Amounts Not Offset in the Consolidated Balance Sheets (D)
	Notes	Gross amounts recognized (A)	Gross amounts offset in the consolidated balance sheets (B)	Net amount presented in the consolidated balance sheets (C=A-B)	Financial instruments	Cash collateral received/ pledged	Net amount (E=C-D)
Trading and hedging derivatives	10, 15	36.349	2.388	33.961	25.020	8.210	731
Reverse repurchase, securities borrowing and similar agreements		35.805	21	35.784	35.618	204	(39)
Total assets		72.154	2.409	69.744	60.637	8.415	692
Trading and hedging derivatives	10, 15	38.693	2.394	36.299	25.020	10.613	667
Repurchase, securities lending and similar agreements		45.977	21	45.956	45.239	420	297
Total liabilities		84.670	2.414	82.256	70.259	11.033	964

December 2018 (Millions of Euros)
					Gross Amounts Not Offset in the Consolidated Balance Sheets (D)
	Notes	Gross amounts recognized (A)	Gross amounts offset in the consolidated balance sheets (B)	Net amount presented in the consolidated balance sheets (C=A-B)	Financial instruments	Cash collateral received/ pledged	Net amount (E=C-D)
Trading and hedging derivatives	10, 15	48.895	16.480	32.415	24.011	7.790	613
Reverse repurchase, securities borrowing and similar agreements		28.074	42	28.032	28.022	169	(159)
Total assets		76.969	16.522	60.447	52.033	7.959	454
Trading and hedging derivatives	10, 15	50.583	17.101	33.481	24.011	6.788	2.682
Repurchase, securities lending and similar agreements		43.035	42	42.993	42.877	34	82
Total liabilities		93.618	17.143	76.474	66.888	6.822	2.765

The amount of recognized financial instruments within derivatives includes the effect in case of compensation with counterparties with which the Group holds netting agreements, while, for repos, it reflects the market value of the collateral associated with the transaction.

7.4. Structural risk

The structural risks are defined, in general terms, as the possibility of sustaining losses due to adverse movements in market risk factors as a result of mismatches in the financial structure of an entity´s balance sheet.

In the Group, the following types of structural risks are defined, according to the nature and the following market factors: interest rate, exchange rate and equity.

The scope of structural risks in the Group is limited to the banking book, excluding market risks in the trading book that are clearly delimited and separated and make up the Market Risks.

The Assets and Liabilities Committee (ALCO) is the main responsible body for the management of structural risks regarding liquidity/ funding interest rate, currency, equity and solvency. Every month, with the participation of the CEO and representatives from the areas of Finance, Risks and Business Areas; this committee monitors the structural risks and is presented with proposals with regard to action plans related with its management for its approval. These management proposals are made by the Finance area with a forward-looking focus, maintaining the alignment with the risk appetite framework, trying to guarantee the recurrence of results and financial stability, as well as to preserve the solvency of the entity. All balance management units have a local ALCO, which is permanently attended by members of the corporate center, and there is a corporate ALCO where management strategies are monitored and presented in the Group's subsidiaries.

GRM area acts as an independent unit, enabling adequate separation between the management and risk control functions, and is responsible for assisting the structural risks in the Group are managed according to the strategy approved by the Board of Directors.

Consequently, GRM deals with the identification, measurement, monitoring and control of those risks and their reporting to the corresponding corporate bodies. Through the Global Risk Management Committee (GRMC), it performs the function of control and risk assessment and is responsible for developing the strategies, policies, procedures and infrastructure necessary to identify, evaluate, measure and manage the significant risks that the BBVA Group faces. To this end, GRM, through the corporate unit of Structural Risks, proposes a scheme of limits and alerts that defines the risk appetite set for each of the relevant structural risk types, both at Group level and by management units, which will be reviewed annually, reporting the situation periodically to the Group's corporate bodies as well as to the GRMC.

Within the three lines of defense scheme in which BBVA's internal control model is established according to the most advanced standards in terms of internal control, the first line of defense is composed by the Finance area, which is responsible for managing the structural risk.

While GRM, as a second line of defense, is in charge of identifying risks, and establishing policies and control models, periodically evaluating their effectiveness.

In the second line of defense, there are also the Internal Risk Control units, which independently review the Structural Risk control, and Internal Financial Control, which carry out a review on the design and effectiveness of the operational controls over structural risk management.

The third line of defense is represented by the Internal Audit area, which, with total independence, is responsible for reviewing specific controls and processes.

7.4.1. Structural interest rate risk

The structural interest-rate risk (“IRRBB”) is related to the potential impact that variations in market interest rates have on an entity's net interest income and equity. In order to properly measure IRRBB, BBVA takes into account the main sources that generate this risk: repricing risk, yield curve risk, option risk and basis risk, which are analyzed with an integral vision, combining two complementary points of view: net interest income (short term) and economic value (long term).

The exposure of a financial entity to adverse interest rates movements is a risk inherent to the development of the banking business, which is also, in turn, an opportunity to create economic value. Therefore, interest rate risk must be effectively managed so that it is limited in accordance with the entity’s equity and in line with the expected economic result.

This function falls to the Global ALM (Asset & Liability Management) unit, within the Finance area, who, through ALCO, aims to support the recurrence of results and preserve the solvency of the entity, always adhering to the risk profile defined by the management bodies of the BBVA Group. The interest rate risk management of the balance sheet aims to promote the stability of the net interest income and book value with respect to changes in market interest rates,types of markets in the different balance-sheets, while respecting solvency and internal limits, as well as complying with current and future regulatory requirements. Likewise, a specific monitoring of the banking book instruments registered at market value (fair value) is developed, which due to their accounting treatment have an impact on results and / or equity.

In this regard, the BBVA Group maintains an exposure to fluctuations on interest rates according to its objective strategy and risk profile, being carried out in a decentralized and independent manner in each of the banking entities that compose its structural balance-sheet.

The management is carried out in accordance with the guidelines established by the European Banking Authority (EBA), with a monitoring of interest rate risk metrics, with the aim of analyzing the potential impact that could be derived from the range of scenarios in the different balance-sheets of the Group.

Nature of Interest Rate Risk

Repricing risk arises due to the difference between the repricing or maturity terms of the assets and liabilities, and represents the most frequent interest rate risk faced by financial entities. However, other sources of risk as changes in the slope and shape of the yield curve, the reference to different indexes and the optionality risk embedded in certain banking transactions, are also taken into account by the risk control system.

BBVA's structural interest-rate risk management process is formed from a set of metrics and tools that enables the capture of additional sources to properly monitor the risk profile of the Group, backed-up by an assumptions set that aims to characterize the behavior of the balance sheet items with the maximum accuracy.

The IRRBB measurement is carried out on a monthly basis, and includes probabilistic measures based on methods of scenario simulation, which enables to capture additional sources of risk to the parallel shifts, as the changes in slope shape and the basis of yield curves. Additionally, sensitivity analysis to multiple parallel shocks of different magnitude are also assessed on a regular basis. The process is run separately for each currency to which the Group is exposed, considering, at a later stage, the diversification effect among currencies and business units.

The risk measurement model is complemented by the assessment of ad-hoc scenarios and stress tests. As stress testing has become more relevant during the recent years, the evaluation of extreme scenarios of rupture of historical interest rates levels, correlations and volatility has continued to be enhanced, while assessing, also, BBVA Research market scenarios, and incorporating the set of scenarios defined according to EBA guidelines.

During 2020, the Group worked to improve the control and management model in accordance with the guidelines established by the EBA on the management of interest rate risk in the banking book. It is worth highlighting, among other aspects, the reinforcement of the stress analysis, including the evaluation of the impacts on the main balance sheet accounts of the Group that could derive from the range of interest rate scenarios defined according to the EBA guidelines mentioned above.

Key assumptions of the model

In order to measure structural interest rate risk, the setting of assumptions on the evolution and behavior of certain balance sheet items is particularly relevant, especially those related to products without an explicit or contractual maturity.

The assumptions that characterize these balance sheet items must be understandable for the areas and bodies involved in risk management and control and remain duly justified and documented. The modeling of these assumptions must be conceptually reasonable and consistent with the evidence based on historical experience, reviewed at least once a year.

In view of the heterogeneity of the financial markets and the availability of historical data, each one of the entities of the Group is responsible for determining the behavior assumptions to be applied to the balance sheet items, always under the guidelines and the applicability of the corporate models existing in the Group.

Among the balance sheet assumptions stand out those established for the treatment of items without contractual maturity, mainly for demand customer deposits, and those related to the expectations on the exercise of interest rate options, especially those relating to loans and deposits subject to prepayment risk.

For the modeling of demand deposits, a segmentation of the accounts in several categories is previously carried out depending on the characteristics of the customer (retail / wholesale) and the product (type of account / transactionality / remuneration), in order to outline the specific behavior of each segment.

In order to establish the remuneration of each segment, the relationship between the evolution of market interest rates and the interest rates of managed accounts is analyzed, with the aim of determining the translation dynamic (percentages and lags) of interest rates variations to the remuneration of the accounts.

The behavior assigned to each category of accounts is determined by an analysis of the historical evolution of the balances and the probability of cancellation of the accounts. For this, the volatile part of the balance assigned to a short-term maturity is isolated, thus avoiding fluctuations in the level of risk caused by specific variations in the balances and promoting stability in the management of the balance. Once the stable part is identified, a medium / long term maturity model is applied through a decay distribution based on the average term of the accounts and the conditional cancellation probabilities throughout the life of the product.

Additionally, the relationship of the evolution of the balance of deposits with the levels of market interest rates is taken into account, where appropriate, including the potential migration between the different types of deposits (on demand / time deposits) in the different interest rate scenarios.

Equally relevant is the treatment of early cancelation options embedded in credit loans, mortgage portfolios and customer deposits. The evolution of market interest rates may condition, along with other variables, the incentive that customers have to prepay loans or deposits, modifying the future behavior of the balance amounts with respect to the forecasted contractual maturity schedule.

The detailed analysis of the historical information related to prepayment data, both partial and total prepayment, combined with other variables such as interest rates, allows estimating future amortizations and, where appropriate, their behavior linked to the evolution of such variables.

The approval and updating of the risk behavior models of structural interest rate risk are subject to corporate governance under the scope of GRM-Analytics. In this way, the models must be properly inventoried and cataloged and comply with the requirements established in the internal procedures for their development, updating and management of the changes. The models are also subject to the corresponding internal validations based on their relevance and the established monitoring requirements.

The table below shows the profile of average interest rate risk in terms of sensitivities of the main currencies in the BBVA Group in 2020:

Sensitivity to interest-rate analysis - December 2020
	Impact on net interest income (*)		Impact on economic value (**)
	100 basis-point increase	100 basis-point decrease (***)	100 basis-point increase	100 basis-point decrease (***)
EUR	[1.5% , 3.5%]	[-1.5% , -0.5%]	[3.5% , 5.5%]	[-3.5% , -1.5%]
MXN	[0.5% , 1.5%]	[-1.5% , -0.5%]	[-1.5% , -0.5%]	[0.5% , 1.5%]
TRY	[-0.5% , 0.5%]	[-0.5% , 0.5%]	[-0.5% , 0.5%]	[-0.5% , 0.5%]
Other	[-0.5% , 0.5%]	[-0.5% , 0.5%]	[-0.5% , 0.5%]	[-0.5% , 0.5%]
BBVA Group	[3.5% , 5.5%]	[-3.5% , -1.5%]	[3.5% , 5.5%]	[-3.5% , -1.5%]

(*) Percentage of "1 year" net interest income forecast for each unit.

(**) Percentage of Core Capital for each unit.

(***) In EUR and USD, negative interest rates scenarios are allowed up to plausible levels lower than current rates.

During 2020, central banks and governments have carried out monetary stimulus measures to mitigate the economic impact caused by the COVID-19 pandemic, which has significantly affected the global economy, spreading to most countries. In Europe, the monetary stimulus measures of the European Central Bank have continued, and the Euribor have fallen, reaching historical low records. In the United States, the reference rates (Libor) have maintained a downward trend, in line with the cuts made by the Federal Reserve in the first quarter of the year. Also in Mexico, the monetary policy rate has fallen significantly during the year. In Turkey, although it initially showed a downward trend in interest rates, aggressive increases have been registered since August, reversing the declines of previous quarters, ending the year with an increase of 500 basis points above December's level of 2019.

In South America, monetary policy has been expansionary, with a reduction in reference rates in the economies of Colombia and Peru, reaching historical low records, affected by the contraction in activity. On the other hand, in Argentina there is a strongly restrictive monetary policy, with a high increase in interest rates in the second half of the year, due to the strong volatility of the markets, affected by the devaluation of the exchange rate.

The BBVA Group, at an aggregate level, continues to maintain a moderate risk profile, in accordance with the established objective, showing a favorable position to a rise in interest rates on net interest income. Effective management of the balance sheet structural risk has mitigated the negative impact of the downward trend in interest rates and the volatility experienced as a result of the effects of COVID-19, and is reflected in the strength and recurrence of the margin of interests:

In Europe and the United States, the downward trend in interest rates remains limited by current levels, preventing extremely adverse scenarios from occurring. Both balance sheets are characterized by a loan portfolio with a high proportion referenced to a variable interest rate (mainly mortgages in Spain and loans to companies in both countries) and a liability composed mainly of customer deposits. The COAP portfolios act as hedging of the bank balance, mitigating its sensitivity to interest rate movements. This profile has remained stable during 2020 on both balance sheets. In Spain, the sensitivity of the interest margin has increased in the year due to the maintenance of higher balances of sensitive liquid assets as a result of the generation of liquidity on the balance sheet and the additional financing of TLTRO III (see Note 22), and due to maturity of a part of the coverage of the mortgage portfolio. In the United States, the sensitivity has been reduced due to the balance sheet hedges carried out in late 2019 and early 2020.

In Mexico, a balance has been maintained between balances referenced to fixed and variable interest rates. Among the assets most sensitive to interest rate movements, the wholesale portfolio stands out, while consumer and mortgages are mostly at a fixed rate. The COAP portfolio is used to balance the longer term of customer deposits. The sensitivity of the interest margin remains limited and stable during 2020, considering the new interest rate scenario that emerged in March, with a downward trend in rates benchmark throughout 2020.

In Turkey, the interest rate risk on the balance sheet increased during 2020, as a result of regulatory requirements (such as the Asset Ratio, applied by the Banking Regulation Supervision Agency (BRSA) and the Good Bank, established by the Central Bank of Turkey (CBRT)) that encourage loan growth. As a result of the establishment of these Regulations, the growth of loans, mostly at a fixed rate, together with the increase in the COAP portfolio, negatively affected sensitivity, being offset by inflation-linked bonds and floating bonds, as well as due to the increase in deposits in the liability side.

In South America, the risk profile on interest rates continues to be low, as most of the countries in the area have a composition of fixed / variable and very similar maturities between assets and liabilities, showing a sensitivity of the margin interest rate limited and with slight variations throughout 2020. Likewise, in countries with balances in several currencies, interest rate risk is also managed for each of the currencies, showing a very low level of risk. The measures promoted by central banks and governments have contributed to raising deposits and excess liquidity in Colombia and Peru, as well as their positions in monetary assets, generating a slight positive variation in margin sensitivity.

7.4.2. Structural exchange-rate risk

Structural exchange rate risk, inherent to the business of international banking groups that develop their activities in different geographies and currencies, is defined as the possibility of impacts on solvency, equity value and results driven by fluctuations in the exchange rates due to exposures in foreign currencies.

In the BBVA Group, structural exchange-rate risk arises from the consolidation of holdings in subsidiaries with functional currencies other than the euro. Its management is centralized in order to optimize the joint management of permanent foreign currency exposures, taking diversification into account.

The corporate Global ALM unit, through ALCO, designs and executes hedging strategies with the main purpose of preserving the stability of consolidated capital ratios and income flows generated in a currency other than the euro in the BBVA Group, keeping a value generation perspective to preserve the Group’s equity in the long term.To this end, a dynamic management strategy is carried out, considering hedge transactions according to market expectations and their costs.

The risk monitoring metrics included in the framework of limits, in line with the Risk Appetite Framework, are integrated into management and supplemented with additional assessment indicators. At the corporate level they are based on probabilistic metrics that measure the maximum deviation in the Group’s Capital, CET1 (“Common Equity Tier 1”) ratio, and net attributable profit. The probabilistic metrics make it possible to estimate the joint impact of exposure to different currencies taking into account the different variability in exchange rates and their correlations.

The suitability of these risk assessment metrics is reviewed on a regular basis through back-testing exercises. The final element of structural exchange-rate risk control is the stress and scenario analysis aimed to assess the vulnerabilities of foreign currency structural exposure not contemplated by the risk metrics and to serve as an additional tool when making management decisions. The scenarios are based both on historical situations simulated by the risk model and on the risk scenarios provided by BBVA Research.

As of December 31, 2020, the main currencies of the geographies where the Group operates have depreciated against the euro during the year: Mexican peso (-13.1%), US Dollar (-8.5%), Turkish lira (-26.7%), Colombian peso (-12.6%), Peruvian sol (-16.3%) and Argentine peso (-34.8%).

The Group's structural exchange-rate risk exposure level has in some cases increased due to the restrictions related to dividend payments from the subsidiaries which have offset the reduction in risk due to the depreciation of the currencies. The hedging policy intends to keep low levels of sensitivity to movements in the exchange rates of emerging markets currencies against the euro. The risk mitigation level in the capital ratio due to the book value of the BBVA Group's holdings in foreign emerging markets currencies stood at around 65% and, as of the end of 2020, CET1 ratio sensitivity to the depreciation of 10% in the euro exchange rate for each currency is estimated: USD +9 bps; Mexican peso -5 bps; Turkish lira -2 bps; other currencies -1 bp (excluding hyperinflation economies). On the other hand, hedging of emerging markets currency denominated earnings in 2020 reached 65%, concentrated in Mexican peso, Turkish lira and the main Latin American currencies.

For the years 2020, 2019 and 2018, the estimated sensitivities of the result attributable to the parent company are shown below, taking into account the coverage against depreciations and appreciations of 1% of the average rate in the main currencies. To the extent that hedging positions are periodically modulated, the sensitivity estimate attempts to reflect an average (or effective) sensitivity in the year:

Sensitivity to 1% (Millions of Euros)
Currency	2020	2019	2018
Mexican peso	4,9	12,7	13,0
Turkish lira	4,5	3,1	3,0
Peruvian sol	0,4	1,9	1,3
Chilean peso	0,3	0,5	0,7
Colombian peso	1,4	2,6	1,9
Argentine peso	0,9	1,3	(0,3)
US Dollar	4,3	5,9	7,3

7.4.3. Structural equity risk

Structural equity risk refers to the possibility of suffering losses in the value of positions in shares and other equity instruments held in the banking book with long or medium term investment horizons due to fluctuations in the value of equity indexes or shares.

BBVA Group's exposure to structural equity risk arises largely from minority shareholdings held on industrial and financial companies. This exposure is modulated in some portfolios with positions held on derivative instruments on the same underlying assets, in order to adjust the portfolio sensitivity to potential changes in equity prices.

The management of structural equity portfolios is a responsibility of Global ALM and other Group's units specialized in this area. Their activity is subject to the risk management corporate policy on structural equity risk management, complying with the defined management principles and Risk Appetite Framework.

The structural equity risk metrics, designed by GRM according to the corporate model, contribute to the effective monitoring of the risk by estimating the sensitivity and the capital necessary to cover the possible unexpected losses due to changes in the value of the shareholdings in the Group's investment portfolio, with a level of confidence that corresponds to the objective rating of the entity, taking into account the liquidity of the positions and the statistical behavior of the assets to be considered

In order to analyze the risk profile in depth, stress tests and scenario analysis of sensitivity to different simulated scenarios are carried out. They are based on both past crisis situations and forecasts made by BBVA Research. These analyses are carried out regularly to assess the vulnerabilities of structural equity exposure not contemplated by the risk metrics and to serve as an additional tool when making management decisions.

Backtesting is carried out on a regular basis on the risk measurement model used.

Global Equity markets have been severely affected by the outbreak of the coronavirus in the first quarter. The extraordinary fiscal and monetary response fostered their recovery although this has been uneven across different geographies and sectors. In this sense, the Spanish equity market has shown one of the worst performances as it fell 15% during 2020.

Structural equity risk, measured in terms of economic capital, has remained fairly stable in the period.The aggregate sensitivity of the BBVA Group’s consolidated equity to a 1% fall in the price of shares of the companies making up the equity portfolio decreased to -€20 million as of December 31, 2020, compared to -€26 million as of December 31, 2019. This estimation takes into account the exposure in shares valued at market prices, or if not applicable, at fair value (excluding the positions in the Treasury Area portfolios) and the net delta-equivalent positions in derivatives on the same underlyings.

7.5. Liquidity and funding risk

Liquidity and funding risk is defined as the incapacity of a bank in meeting its payment commitments due to lack of funds or that, to face those commitments, should have to make use of funding under burdensome terms.

7.5.1. Liquidity and Funding Strategy and Planning

The BBVA Group is a multinational financial institution whose business is focused mainly on retail and commercial banking activities. In addition to the retail business model, which forms its core business, the Group engages in corporate and investment banking, through the global CIB (Corporate & Investment Banking) division.

Liquidity and funding risk management aims to maintain a solid balance sheet structure which allows a sustainable business model. The Group’s liquidity and funding strategy is based on the following pillars:

The principle of the funding self-sufficiency of its subsidiaries, meaning that each of the Liquidity Management Units (LMUs) must cover its funding needs independently on the markets where it operates. This avoids possible contagion due to a crisis affecting one or more of the Group’s LMUs.

Stable customer deposits as the main source of funding in all the LMUs, in accordance with the Group’s business model.

Diversification of the sources of wholesale funding, in terms of maturity, market, instruments, counterparties and currencies, with recurring access to the markets.

Compliance with regulatory requirements, enabling the availability of ample liquidity buffers, of high quality, as well as sufficient instruments as required by regulations with the capacity to absorb losses.

Compliance with the internal Liquidity Risk and Funding metrics, while adhering to the Risk Appetite level established for each LMU at any time.

Liquidity and Financing Risk Management aims, in the short term, to prevent an entity from having difficulties in meeting its payment commitments in due time and form or that, to meet them, it has to resort to obtaining funds in burdensome conditions that deteriorate the image or reputation of the entity.

In the medium term, its objective is to support the suitability of the Group's financial structure and its evolution, within the framework of the economic situation, the markets and regulatory changes.

This management of structural and liquidity funding is based on the principle of financial self-sufficiency of the entities that comprise it. This approach helps prevent and limit liquidity risk by reducing the Group’s vulnerability during periods of high risk. This decentralized management prevents possible contagion from a crisis affecting only one or a few Group entities, which must act independently to meet their liquidity requirements in the markets where they operate.

Within this strategy, the BBVA Group is organized into eleven LMUs compose of the parent company and the bank subsidiaries in each geography, plus the branches that depend on them.

In addition, the policy for managing liquidity and funding risk is also based on the model’s robustness and on the planning and integration of risk management into the budgeting process of each LMU, according to the financing risk appetite that it decides to assume in its business.

Liquidity and funding planning is part of the strategic processes for the Group’s budgetary and business planning. This objective is to allow a recurrent growth of the banking business with suitable maturities and costs within the established risk tolerance levels by using a wide range of instruments which allow the diversification of the funding sources and the maintenance of a high volume of available liquid assets.

7.5.2. Governance and monitoring

The responsibility for liquidity and funding management in the development of normal business activity lies with the Finance area as a first line of defense in managing the risks inherent to this activity, in accordance with the principles established by the European Banking Authority (EBA) and in line with the most demanding standards, policies, procedures and controls in the framework established by the governing bodies. Finance, through the Balance-Sheet Management area, plans and executes the funding of the structural long-term gap of each LMU and proposes to the Assets and Liabilities Committee (ALCO) the actions to be taken on this matter, in accordance with the policies established by the Risk Committee in line with the metrics of the Risk Appetite Framework approved by the Board of Directors.

Finance is also responsible for preparing the regulatory reporting of liquidity, coordinating with the responsible areas in each LGU the necessary processes to cover the requirements at corporate and regulatory level, enabling the integrity of the information provided.

GRM is responsible for promoting the liquidity and financing risk in the Group is managed in accordance with the framework established by governing bodies. It also deals with the identification, measurement, monitoring and control of such risks and their communication to the relevant corporate bodies. In order to carry out this task properly, the risk function in the Group has been configured as a single, global function, independent of the management areas.

Additionally, the Group has, in its second line of defense, an Internal Risk Control unit, which performs an independent review of the control of Liquidity and Financing Risk, and a Financial Internal Control Unit that reviews the design and effectiveness of the controls operations on liquidity management and reporting.

As the third line of defense of the Group's internal control model, Internal Audit is in charge of reviewing specific controls and processes in accordance with a work plan that is drawn up annually.

The Group’s fundamental objectives regarding the liquidity and funding risk are determined through the Liquidity Coverage Ratio (LCR) and through the Loan-to-Stable Customer Deposits (LtSCD) ratio.

The LCR ratio is a regulatory metric that aims to enable the resilience of entities in a scenario of liquidity tension within a time horizon of 30 days. Within its risk appetite framework and system of limits and alerts, BBVA has established a required LCR compliance level for the entire Group and for each individual LMU. The internal levels required are aimed at efficiently meeting the regulatory requirement, at a widely level above 100%.

The LtSCD ratio measures the relationship between net lending and stable customer funds. The aim is to preserve a stable funding structure in the medium term for each of the LMUs which make up the BBVA Group, taking into account that maintaining an adequate volume of stable customer funds is key to achieving a sound liquidity profile. In geographical areas with dual-currency balances, the indicator is also controlled by currency to manage the mismatches that might occur.

Stable customer funds are considered to be the financing obtained and managed from the LMUs among their target customers. Those funds are characterized by their low sensitivity to market changes and by their less volatile behavior at aggregated level per operation due to the loyalty of the customer to the entity. The stable resources are calculated by applying to each identified customer segment a haircut determined by the analysis of the stability if the balances by which different aspects are evaluated (concentration, stability, level of loyalty). The main source of stable resources arises from wholesale funding and retail customer funds.

In order to establish the target (maximum) levels of LtSCD in each LMU and provide an optimal funding structure reference in terms of risk appetite, the corporate Structural Risks unit of GRM identifies and assesses the economic and financial variables that condition the funding structures in the different geographical areas.

Additionally, liquidity and funding risk management aims to achieve a proper diversification of the funding structure, avoiding excessive dependence on short-term funding by establishing a maximum level for the short-term funds raised, including both wholesale financing and the least stable proportion of customer funds In relation to long-term financing, the maturity profile does not present significant concentrations, which makes it possible to adapt the schedule of the planned issuance plan to the best financial conditions in the markets. Lastly, concentration risk is monitored at LMU level, with the aim of supporting a correct diversification of both the counterparty and type of instrument.

One of the fundamental metrics within the general management framework of the liquidity and funding risk is the maintenance of a liquidity buffer consisting of high quality assets free of charges which can be sold or offered as collateral to obtain funding, either under normal market conditions or in stress situations.

The Finance is responsible for the collateral management and determining the liquidity buffer within the BBVA Group. According to the principle of auto-sufficiency of the Group's subsidiaries, each LMU is responsible for maintaining a buffer of liquid assets which complies with the regulatory requirements applicable under each jurisdiction. In addition, the liquidity buffer of each LMU must be aligned with the liquidity and funding risk tolerance as well as the management limits set and approved for each case.

In this context, the short-term resistance of the liquidity risk profile is promoted, supporting that each LMU has sufficient collateral to deal with the risk of the closure of wholesale markets. Basic capacity is the internal metric for the management and control of short-term liquidity risk, which is defined as the relationship between the explicit assets available and the maturities of wholesale liabilities and volatile resources, at different time periods up to the year, with special relevance at 30 and 90 days, with the objective of preserving the survival period above 3 months with the available buffer, without considering the balance inflows.

As a fundamental element of the liquidity and financing risk monitoring scheme, stress tests are carried out. They enable to anticipate deviations from the liquidity targets and the limits set in the appetite, and to establish tolerance ranges in the different management areas. They also play a major role in the design of the Liquidity Contingency Plan and the definition of specific measures to be adopted to rectify the risk profile if necessary.

For each scenario, it is checked whether BBVA has a sufficient stock of liquid assets to assist its capacity to meet the liquidity commitments/outflows in the different periods analyzed. The analysis considers four scenarios: one central and three crisis-related (systemic crisis; unexpected internal crisis with a considerable rating downgrade and/or affecting the ability to issue in wholesale markets and the perception of business risk by the banking intermediaries and the entity’s clients; and a mixed scenario, as a combination of the two aforementioned scenarios). Each scenario considers the following factors: existing market liquidity, customer behavior and sources of funding, the impact of rating downgrades, market values of liquid assets and collateral, and the interaction between liquidity requirements and the development of BBVA's credit quality.

The stress tests conducted on a regular basis by GRM reveal that BBVA maintains a sufficient buffer of liquid assets to deal with the estimated liquidity outflows in a scenario resulting from the combination of a systemic crisis and an unexpected internal crisis, during a period of longer than 3 months in general for the different LMUs (with the exception of Turkey where despite closing the year above 3 months, the regulatory requirements have led to non-compliance during certain periods), including in the scenario of a significant downgrade of the Bank’s rating by up to three notches.

Together with the results of the stress tests and the risk metrics, the early warning indicators play an important role within the corporate model and the Liquidity Contingency Plan. They are mainly indicators of the funding structure, in relation to asset encumbrance, counterparty concentration, flights of customer deposits, unexpected use of credit facilities, and of the market, which help anticipate possible risks and capture market expectations.

Finance is the area responsible for the elaboration, monitoring, execution and update of the liquidity and funding plan and of the market access strategy to assist and improve the stability and diversification of the wholesale funding sources.

In order to implement and establish management in an anticipated manner, limits are set on an annual basis for the main management metrics that form part of the budgeting process for the liquidity and funding plan. This framework of limits contributes to the planning of the joint future performance of:

The loan book, considering the types of assets and their degree of liquidity, as well as their validity as collateral in collateralized funding.

Stable customer funds, based on the application of a methodology for establishing which segments and customer balances are considered to be stable or volatile funds based on the principle of sustainability and recurrence of these funds.

Projection of the credit gap, in order to require a degree of self-funding that is defined in terms of the difference between the loan-book and stable customer funds.

Incorporating the planning of securities portfolios into the banking book, which include both fixed-interest and equity securities, and are classified as financial assets at fair value through other comprehensive income and at amortized cost, and additionally on trading portfolios.

The structural gap projection, as a result of assessing the funding needs generated both from the credit gap and by the securities portfolio in the banking book, together with the rest of on-balance-sheet wholesale funding needs, excluding trading portfolios. This gap therefore needs to be funded with customer funds that are not considered stable or on wholesale markets.

As a result of these funding needs, the BBVA Group plans the target wholesale funding structure according to the tolerance set in each LMU target.

Thus, once the structural gap has been identified and after resorting to wholesale markets, the amount and composition of wholesale structural funding is established in subsequent years, in order to maintain a diversified funding mix and guarantee that there is not a high reliance on short-term funding (short-term wholesale funding plus volatile customer funds).

In practice, the execution of the principles of planning and self-funding at the different LMUs results in the Group’s main source of funding being customer deposits, which consist mainly of demand deposits, savings deposits and time deposits.

As sources of funding, customer deposits are complemented by access to the interbank market and the domestic and international capital markets in order to address additional liquidity requirements, implementing domestic and international programs for the issuance of commercial paper and medium and long-term debt.

The process of analysis and assessment of the liquidity and funding situation and of the inherent risks is a process carried out on an ongoing basis in the BBVA Group, with the participation of all the Group areas involved in liquidity and funding risk management. This process is carried out at both local and corporate level. It is incorporated into the decision- making process for liquidity and funding management, with integration between the risk appetite strategy and establishment and the planning process, the funding plan and the limits scheme.

7.5.3. Liquidity and funding performance

During 2020, the BBVA Group has maintained a robust and dynamic funding structure with a predominantly retail nature, where customer resources represent the main source of funding.

During 2020, liquidity conditions have remained comfortable in all the countries where the BBVA Group operates. Since the beginning of March, the global crisis caused by COVID-19 has had a significant impact on financial markets. The effects of this crisis on the Group's balance sheets materialized fundamentally at first, through greater provision of credit lines by wholesale clients in view of the worsening financing conditions in the markets, with no significant effect on the retail world. These provisions were largely paid off over the following quarters. Dealing with this situation of initial uncertainty, the different central banks provided a joint response through specific measures and programs to facilitate the financing of the real economy and the provision of liquidity in financial markets, increasing liquidity buffers in almost all areas with BBVA presence.

Thus, the performance of the indicators show that the robustness of the funding structure remained steady during 2020, 2019 and 2018, in the sense that all LMUs held self-funding levels with stable customer resources above the requirements.

LtSCD by LMU
	2020	2019	2018
Group (average)	95%	108%	106%
Eurozone	97%	108%	101%
BBVA USA	92%	111%	119%
BBVA Mexico	98%	116%	114%
Garanti BBVA	95%	99%	110%
Other LMUs	86%	103%	99%

With respect to LCR, the Group has maintained a liquidity buffer at both a consolidated and individual level in 2020. As a result, the ratio has remained comfortably above 100%, with the consolidated ratio as of December 31, 2020 standing at 149%.

Although this requirement is only established at a Group level, for banks in the Eurozone, the minimum level required is comfortably exceeded in all subsidiaries. It should be noted that the calculation of the Consolidated LCR does not allow the transfer of liquidity between subsidiaries, so no excess liquidity may be transferred from these entities for the purpose of calculating the consolidated ratio. If the impact of these highly liquid assets was considered, the LCR would be 185%, or +36 basis points above the required level.

LCR main LMU
0	2020	2019	2018
Group	149%	129%	127%
Eurozone	173%	147%	145%
BBVA USA (*)	144%	145%	143%
BBVA Mexico	196%	147%	154%
Garanti BBVA	183%	206%	209%

(*) BBVA USA LCR calculated according to local regulation (Fed Modified LCR).

Each entity maintains an individual liquidity buffer, both BBVA, S.A. and each of its subsidiaries, including BBVA USA, BBVA Mexico, Garanti BBVA and the Latin American subsidiaries.

The table below shows the liquidity available by instrument as of December 31, 2020, 2019 and 2018 for the most significant entities based on prudential supervisor’s information (Commission Implementing Regulations (EU) 2017/2114 of November 9, 2017):

December 2020 (Millions of Euros)
	BBVA Eurozone	BBVA Mexico	Garanti BBVA	Other
Cash and withdrawable central bank reserves	39.330	8.930	6.153	6.831
Level 1 tradable assets	48.858	9.205	7.019	6.237
Level 2A tradable assets	5.119	106	-	-
Level 2B tradable assets	6.080	11	-	-
Other tradable assets	20.174	421	701	745
Cumulated counterbalancing capacity	119.560	18.672	13.873	13.814

December 2019 (Millions of Euros)
	BBVA Eurozone	BBVA Mexico	BBVA USA	Garanti BBVA	Other
Cash and withdrawable central bank reserves	14.516	6.246	4.949	6.450	6.368
Level 1 tradable assets	41.961	7.295	11.337	7.953	3.593
Level 2A tradable assets	403	316	344	-	-
Level 2B tradable assets	5.196	219	-	-	12
Other tradable assets	22.213	1.269	952	669	586
Non tradable assets eligible for central banks	-	-	2.935	-	-
Cumulated counterbalancing capacity	84.288	15.344	20.516	15.072	10.559

December 2018 (Millions of Euros)
	BBVA Eurozone	BBVA Mexico	BBVA USA	Garanti BBVA	Other
Cash and withdrawable central bank reserves	26.506	7.666	1.667	7.633	6.677
Level 1 tradable assets	29.938	4.995	10.490	6.502	3.652
Level 2A tradable assets	449	409	510	-	-
Level 2B tradable assets	4.040	33	-	-	-
Other tradable assets	8.772	1.372	1.043	499	617
Non tradable assets eligible for central banks	-	-	2.314	-	-
Cumulated counterbalancing capacity	69.705	14.475	16.024	14.634	10.946

The Net Stable Funding Ratio (NSFR), defined as the ratio between the amount of stable funding available and the amount of stable funding required, is one of the Basel Committee's essential reforms, and requires banks to maintain a stable funding profile in relation to the composition of their assets and off-balance-sheet activities. This ratio should be at least 100% at all times.

The NSFR of BBVA Group and its main LMUs at December 31, 2020 and 2019, calculated based on the Basel requirements, was the following:

NSFR main LMU
	2020	2019
Group	127%	120%
BBVA Eurozone	121%	113%
BBVA Mexico	138%	130%
BBVA USA	126%	116%
Garanti BBVA	154%	151%

Below is a matrix of residual maturities by contractual periods based on supervisory prudential reporting as of December 31, 2020, 2019 and 2018:

December 2020. Contractual maturities (Millions of Euros)
ASSETS
Cash, cash balances at central banks and other demand deposits	42.518	32.741	-	-	-	-	-	-	-	-	75.258
Deposits in credit entities	-	3.616	677	921	356	461	117	120	2	39	6.309
Deposits in other financial institutions	-	2.202	855	797	734	543	1.251	721	515	500	8.119
Reverse repo, securities borrowing and margin lending	-	20.033	4.757	1.351	364	368	3.320	1.849	891	1.089	34.021
Loans and advances	279	16.939	24.280	23.012	15.579	17.032	46.182	38.851	51.709	110.173	344.036
Securities' portfolio settlement	-	3.896	6.680	6.557	5.084	13.014	9.858	15.494	17.231	50.045	127.859

December 2020. Contractual maturities (Millions of Euros)
LIABILITIES
Wholesale funding	-	4.750	2.618	3.963	1.283	1.543	10.573	7.505	12.793	23.839	68.868
Deposits in financial institutions	8.838	7.859	254	741	152	726	825	189	166	371	20.120
Deposits in other financial institutions and international agencies	12.735	4.324	2.694	588	353	272	957	337	459	870	23.589
Customer deposits	308.360	39.978	13.416	6.808	4.526	4.366	3.361	1.213	869	799	383.694
Security pledge funding	-	41.239	5.301	1.643	1.192	368	11.304	28.510	3.740	1.516	94.812
Derivatives, net	-	(722)	15	(961)	(85)	134	(400)	(157)	(264)	(159)	(2.599)

December 2019. Contractual maturities (Millions of Euros)
	Demand	Up to 1 month	1 to 3 months	3 to 6 months	6 to 9 months	9 to 12 months	1 to 2 years	2 to 3 years	3 to 5 years	Over 5 years	Total
ASSETS
Cash, cash balances at central banks and other demand deposits	20.954	20.654	-	-	-	-	-	-	-	-	41.608
Deposits in credit entities	-	3.591	283	488	585	503	189	24	120	432	6.216
Deposits in other financial institutions	-	1.336	1.120	796	589	991	1.420	1.072	672	2.089	10.084
Reverse repo, securities borrowing and margin lending	-	21.612	3.858	2.287	561	808	4.121	1.838	411	803	36.299
Loans and advances	157	22.015	25.056	24.994	15.777	16.404	42.165	35.917	54.772	122.098	359.354
Securities' portfolio settlement	-	1.622	3.873	6.620	2.017	7.292	21.334	6.115	13.240	46.022	108.136

December 2019. Contractual maturities (Millions of Euros)
	Demand	Up to 1 month	1 to 3 months	3 to 6 months	6 to 9 months	9 to 12 months	1 to 2 years	2 to 3 years	3 to 5 years	Over 5 years	Total
LIABILITIES
Wholesale funding	1	1.393	1.714	4.208	1.645	4.386	8.328	10.608	10.803	27.840	70.927
Deposits in financial institutions	7.377	7.608	493	1.122	172	1.514	386	614	206	510	20.004
Deposits in other financial institutions and international agencies	10.177	3.859	867	381	367	257	982	503	499	952	18.843
Customer deposits	271.638	43.577	18.550	10.013	7.266	6.605	3.717	2.062	854	1.039	365.321
Security pledge funding	-	45.135	3.202	15.801	1.456	653	3.393	7.206	759	1.308	78.914
Derivatives, net	-	(66)	(25)	29	(11)	1.097	(830)	(278)	(333)	(420)	(838)

December 2018. Contractual maturities (Millions of Euros)
	Demand	Up to 1 month	1 to 3 months	3 to 6 months	6 to 9 months	9 to 12 months	1 to 2 years	2 to 3 years	3 to 5 years	Over 5 years	Total
ASSETS
Cash, cash balances at central banks and other demand deposits	9.550	40.599	-	-	-	-	-	-	-	-	50.149
Deposits in credit entities	801	3.211	216	141	83	152	133	178	27	1.269	6.211
Deposits in other financial institutions	1	1.408	750	664	647	375	1.724	896	1.286	2.764	10.515
Reverse repo, securities borrowing and margin lending	-	21.266	1.655	1.158	805	498	205	1.352	390	210	27.539
Loans and advances	132	19.825	25.939	23.265	15.347	16.433	42.100	32.336	53.386	120.571	349.334
Securities' portfolio settlement	-	1.875	4.379	5.990	2.148	6.823	8.592	12.423	11.533	42.738	96.501

December 2018. Contractual maturities (Millions of Euros)
	Demand	Up to 1 month	1 to 3 months	3 to 6 months	6 to 9 months	9 to 12 months	1 to 2 years	2 to 3 years	3 to 5 years	Over 5 years	Total
LIABILITIES
Wholesale funding	1	2.678	1.652	2.160	2.425	2.736	7.225	8.578	16.040	26.363	69.858
Deposits in financial institutions	7.107	5.599	751	1.992	377	1.240	1.149	229	196	904	19.544
Deposits in other financial institutions and international agencies	10.680	4.327	1.580	458	302	309	781	304	825	1.692	21.258
Customer deposits	252.630	44.866	18.514	10.625	6.217	7.345	5.667	2.137	1.207	1.310	350.518
Security pledge funding	40	46.489	2.219	2.274	114	97	22.911	526	218	1.627	76.515
Derivatives, net	-	(75)	(523)	(68)	(5)	(117)	498	(91)	(67)	(392)	(840)

With regard to the financing structure, the loan portfolio is mostly financed by retail deposits. The “demand” maturity bucket mainly contains the retail customer sight accounts whose behavior historically showed a high level of stability and little concentration. According to a behavior analysis which is done every year in every entity, this type of account is considered to be stable and for liquidity risk purposes receive a better treatment.

The liquidity situation of the Group's main management units is detailed below:

In the Euro Liquidity Management Unit (UGL), the liquidity and financing situation remains solid and comfortable with a large high-quality liquidity buffer that has been increased during the year as a result of the growth in customer deposits and the actions taken by the European Central Bank, which have meant an injection of liquidity in the system. As a result of the COVID-19 crisis, there was initially a greater demand for credit through the increase in the use of credit lines by the Corporate & Investment Banking wholesale business, which was also accompanied by a growth in customer deposits. Subsequently, there were partial refunds of those lines while deposits have continued to grow. In addition, it is important to note the measures implemented by the ECB to deal with this crisis, which have included different actions such as: the expansion of asset purchase programs, especially through the PEPP (Pandemic Emergency Purchase Program) for 750,000 million of euros in a first tranche announced in March and expanded with a second tranche for an additional 600,000 million euros until June 2021 or until the ECB considers that the crisis has ended, the coordinated action of central banks for the provision of US dollars, a temporary package of measures to make flexible the collateral eligible for financing operations, the relaxation and improvement of the conditions of the TLTRO III program and the creation of the new program of long-term refinancing operations without specific emergency objective (PELTRO). In this regard, BBVA attended the TLTRO III program windows in March and June (with an amount drawn down at the end of December of 35,032 million euros) due to its favorable conditions in terms of cost and term, amortizing the corresponding part of the TLTRO II program (see Note 22).

In the United States there is a comfortable liquidity situation with significant growth in deposits during the year, driven mainly by stimulus measures from the American government and the Federal Reserve. This has led to an increase in the liquidity buffer and the liquidity and financing indicators are comfortable. As in the euro zone, during the end of the first quarter of 2020 there was an increase in loans stemmed mainly from an increase in the use of credit lines by wholesale clients and the stimulus program of the American government aimed at SMEs and freelances (Paycheck Protection Program). Subsequently, there were repayments that bring the percentage of use of credit lines to levels prior to the pandemic.

In Mexico, the liquidity position has remained solid during the year due to the increase in deposits driven by the success of the commercial actions carried out by the entity, especially in the second semester, as well as by the stimulus measures implemented by Banxico throughout the year to provide liquidity to the financial system, which made it possible to offset the increase in the use of credit lines as a result of the COVID-19 crisis. This good performance in deposits, together with the normalization in credit growth, has reduced the credit gap, resulting in the entity being in a comfortable situation in liquidity and financing ratios.

At Garanti BBVA, the liquidity situation remained comfortable during 2020, with a contraction of loans and a growth of deposits in foreign currency, as well as a higher growth in loans than deposits in local currency. As a result of the COVID-19 crisis, the Turkish regulator established the so-called asset ratio to mainly increase loans and discourage the accumulation of deposits, causing an increase in the credit gap, which was covered with the excess liquidity that the entity had. Subsequently, the asset ratio requirement was reduced in the third quarter (from 100% to 90%) and was eliminated in December. All in all, Garanti BBVA has shown a solid liquidity buffer.

In South America, an adequate liquidity situation is maintained throughout the region, favored by the support of the different central banks and governments that, with the aim of mitigating the impact of the COVID-19 crisis, have implemented measures for stimulating economic activity and providing greater liquidity to financial systems. In Argentina, the outflow of deposits in US dollars in the banking system slowed down during 2020, and even showed some growth in the fourth quarter. BBVA Argentina continues to maintain a solid liquidity position BBVA Colombia, after the actions carried out to adjusting excess liquidity by reducing wholesale deposits, continues to show a comfortable liquidity position. BBVA Peru has seen its comfortable liquidity situation strengthened as a result of the continuous increase in the volume of deposits during the second semester, as well as the funds from the Central Bank's support programs.

The wholesale financing markets in which the Group operates, after the first two months of 2020 of great stability were followed by a strong correction derived from the COVID-19 crisis and limited access to the primary market. This situation has been stabilizing due to the evolution of the pandemic, the development of vaccines, various geopolitical events and the actions of the Central Banks. Secondary market volumes ended the year reaching the levels of January 2020, while primary market volumes have been reactivated, lowering the issue premiums.

The main transactions carried out by the companies that form part of the BBVA Group in 2020 were:

During the first quarter of 2020 BBVA, S.A. made 2 senior non-preferred securities issues for a total of 1,400 million euros and another Tier 2 for 1,000 million euros. In the second quarter of 2020, an issuance of senior preferred securities for 1,000 million euros was executed as a social-COVID-19 bond, the first of its kind for a private financial entity in Europe. In the third quarter, three public issues were made: the first is the first green convertible bond of a financial institution world-wide for an amount of 1,000 million euros; the second is a Tier 2 subordinated securities issue denominated in pound sterling, for an amount of 300 million pounds; and the third is an issuance of preferred securities registered with the US SEC (Securities Exchange Commission) in two tranches with maturities of three and five years, for a total of 2,000 million dollars. On the other hand, in February 2020, BBVA exercised the call option of a convertible bond of 1,500 million euros, and in January 2021, the entity has early amortized three preferred issuances (for more information on these transactions see the section “Solvency” of this report).

In 2020, BBVA México successfully carried out a local senior issuance of 15,000 million Mexican pesos (614 million euros) in three tranches (two tranches in Mexican pesos at 3 and 5 years and another tranche in US dollars at 3 years), in order to advance the refinancing of maturities in the year taking advantage of the good market moment. It also carried out an international issue of senior unsecured securities for an amount of 500 million US dollars of 5 years with a rate of 1.875%, which represents the lowest in history for a financial institution in Mexico and for any private financial institutions in Latin America. Furthermore, within the measures adopted by Banxico throughout the year, BBVA Mexico has participated in auctions of US dollars with credit institutions (swap line with the Fed) initially for an amount of 1,250 million US dollars, partially renewing that position from June to September, for an amount of US $ 700 million. Likewise, it has participated in the so-called Banxico facilities 7 and 8 (measures to transfer funds to micro, small and medium-sized companies, as well as to individuals affected by the pandemic).

In Turkey, Garanti BBVA carried out a Tier 2 issuance for TRY 750 million in the first quarter of 2020. In the second quarter of 2020, Garanti BBVA renewed a syndicated loan by issuing the first green syndicated loan indexed to sustainability criteria, and in whose renovation the EBRD -European Bank for Reconstruction and Development- and the IFC -International Finance Corporation- have participated. And in the fourth quarter, Garanti BBVA partially renewed a syndicated loan for an amount of $636 million.

The liquidity position of the rest of subsidiaries has continued to be sound, maintaining a solid liquidity position in all the jurisdictions in which the Group operates.

In this context, BBVA has maintained its objective of strengthening the funding structure of the different Group entities based on growing their self-funding from stable customer funds, while enabling a sufficient buffer of fully available liquid assets, diversifying the various sources of funding available, and optimizing the generation of collateral available for dealing with stress situations in the markets.

7.5.4. Asset encumbrance

As of December 31, 2020, 2019 and 2018, the encumbered (those provided as collateral for certain liabilities) and unencumbered assets are broken down as follows:

December 2020 (Millions of Euros)
	Encumbered assets		Non-encumbered assets
	Book value	Market value	Book value	Market value
Assets	121.999		614.260
Equity instruments	2.134	2.134	14.556	14.556
Debt securities	29.379	26.112	100.108	100.108
Loans and advances and other assets	90.486		499.595

December 2019 (Millions of Euros)
	Encumbered assets		Non-encumbered assets
	Book value	Market value	Book value	Market value
Assets	101.792		596.898
Equity instruments	3.526	3.526	12.113	12.113
Debt Securities	29.630	29.567	95.611	95.611
Loans and Advances and other assets	68.636	-	489.174	-

December 2018 (Millions of Euros)
	Encumbered assets		Non-encumbered assets
	Book value	Market value	Book value	Market value
Assets	107.950		567.573
Equity instruments	1.864	1.864	6.485	6.485
Debt Securities	31.157	32.216	82.209	82.209
Loans and Advances and other assets	74.928		478.880

The committed value of "Loans and Advances and other assets" corresponds mainly to loans linked to the issue of covered bonds, territorial bonds or long-term securitized bonds (see Note 22.4) as well as those used as a guarantee to access certain funding transactions with central banks. Debt securities and equity instruments correspond to underlying that are delivered in repos with different types of counterparties, mainly clearing houses or credit institutions, and to a lesser extent central banks. Collateral provided to enablederivative transactions is also included as committed assets.

As of December 31, 2020, 2019 and 2018, collateral pledges received mainly due to repurchase agreements and securities lending, and those which could be committed in order to obtain funding are provided below:

December 2020. Collateral received (Millions of Euros)
Collateral received	30.723	8.652	1.071
Equity instruments	239	204	-
Debt securities	30.484	8.448	1.071
Own debt securities issued other than own covered bonds or ABSs	3	94	-

December 2019. Collateral received (Millions of Euros)
Collateral received	38.496	9.208	48
Equity instruments	65	70	-
Debt securities	38.431	9.130	38
Loans and advances and other assets	-	8	10
Own debt securities issued other than own covered bonds or ABSs	-	82	-

December 2018. Collateral received (Millions of Euros)
	Fair value of encumbered collateral received or own debt securities issued	Fair value of collateral received or own debt securities issued available for encumbrance	Nominal amount of collateral received or own debt securities issued not available for encumbrance
Collateral received	27.474	5.633	319
Equity instruments	89	82	-
Debt securities	27.385	5.542	300
Loans and Advances and other assets	-	8	19
Own debt securities issued other than own covered bonds or ABSs	78	87	-

The guarantees received in the form of reverse repurchase agreements or security lending transactions are committed by their use in repurchase agreements, as is the case with debt securities.

As of December 31, 2020, 2019 and 2018, financial liabilities issued related to encumbered assets in financial transactions as well as their book value were as follows:

Sources of encumbrance (Millions of Euros)
	2020		2019		2018
	Matching liabilities, contingent liabilities or securities lent	Assets, collateral received and owndebt securities issued other than covered bonds and ABSs encumbered	Matching liabilities, contingent liabilities or securities lent	Assets, collateral received and owndebt securities issued other than covered bonds and ABSs encumbered	Matching liabilities, contingent liabilities or securities lent	Assets, collateral received and owndebt securities issued other than covered bonds and ABSs encumbered
Book value of financial liabilities	131.352	147.523	124.252	135.500	113.498	131.172
Derivatives	16.611	16.348	19.066	20.004	8.972	11.036
Loans and advances	98.668	111.726	87.906	94.240	85.989	97.361
Outstanding subordinated debt	16.073	19.449	17.280	21.256	18.538	22.775
Other sources	653	5.202	449	4.788	3.972	4.330