CORRESP 1 filename1.htm


July 29, 2020



Securities and Exchange Commission
Division of Corporation Finance
Office of Manufacturing
100 F Street, N.E.
Washington, D.C.  20549

Attention: Jeffrey Gordon and Anne McConnell, Division of Corporation Finance Office of Manufacturing

Form 20-F for the Year ended December 31, 2019
Filed March 3, 2020
File No. 1-35788

Dear Mr. Gordon and Ms. McConnell:

By letter dated July 15, 2020, the Division of Corporation Finance Office of Manufacturing (the “DCFOM”) of the Securities and Exchange Commission (the “Commission”) sent ArcelorMittal (“ArcelorMittal” or the “Company”) comments on its above-referenced Annual Report on Form 20-F (“Form 20-F”). This letter sets forth ArcelorMittal’s responses to the DCFOM’s comments, the text of each of which we have incorporated into this response letter for convenience.



Securities and Exchange Commission, p. 2

We note the company’s net book value significantly exceeds its market capitalization. We also note this trend has persisted for a considerable period of time and the degree to which net book value exceeds market capitalization has substantially increased. Given that declines in market capitalization may indicate potential asset impairments, please explain if and how this factor has impacted your asset impairment assessments. Please more fully explain and address the following:

·Explain how you determine your cash generating units for impairment testing of property, plant and equipment;
·Explain how you determine and assess the reasonableness of the useful lives of property, plant and equipment;
·Explain how you determine your groups of cash generating units for impairment testing of goodwill;
·Since recoverable amounts appear to exceed market capitalization, explain how you determine the material assumptions underlying your impairment analyses and how you determine each material assumption is reasonable and supportable; and
·Explain how you assess the reasonableness of your fair value estimates, including if and how you reconcile your fair value estimates to your market capitalization.

The Company acknowledges the DCFOM’s comments and provides the following responses.


Impairment assessment considering market capitalization


In assessing whether there is any indication that an asset may be impaired, the Company considers the external and internal sources of information in accordance with paragraph 12 and 14 of IAS 36 Impairment of Assets.

In accordance with paragraph 12(d) of IAS 36, the Company considers the evolution of its market capitalization in the overall assessment of impairment indicators at each reporting date. The property, plant and equipment (“PP&E”) of all cash-generating units (“CGU”) is tested as part of the annual impairment test of goodwill. 

The Company considers the below factors regarding market capitalization:

·The market may have taken account of factors other than the return that the entity is generating on its assets, such as short-term factors related to the macroeconomic, political or social environment and which may cause significant volatility in the share price. As an example, in recent history in the context of the COVID-19 pandemic, ArcelorMittal’s share price decreased by 61% between early February and the end of March but then subsequently increased again by 57%. ArcelorMittal’s share price tends to fluctuate based on any expected impacts on near-term financial performance. In the current context, ArcelorMittal shares are trading at a multiple of expected 2020/21 earnings, rather than the Company’s medium-term earnings potential. Therefore, the Company does not believe that the share price fully reflects the medium and long-term earning expectations of the Company.


Securities and Exchange Commission, p.3

ArcelorMittal expects its share price to recover to a higher level when its operating environment normalizes and the market horizon shifts forward.

·Being a capital-intensive industry with significant plant and machinery equipment where assets are operated over tens of years of useful life and investment decisions are conditioned by long-term returns, the value of the steel industry is less likely to be impacted by short-term market swings, which do not to capture its long-term value.
·ArcelorMittal’s stock is cyclical as described in various analyst reports and the market traditionally de-rates but also re-rates such stocks in favour of “growth” stocks in particular when there are severe short-term macroeconomic fluctuations.

Accordingly, the Company concluded that its CGUs and groups of cash-generating units (“GCGUs”) from a goodwill impairment testing perspective are much more sensitive to changes such as a decline in value in use (“VIU”) due to the passage of time in normal use; changes in the technological, market or economic environment; physical damage; or indication of future long-term performance being lower than initially expected.


Determination of cash generating units for impairment testing of PP&E


As described in note 5.3 to the consolidated financial statements as of and for the year ended December 31, 2019 included in the Form 20-F (the “2019 Consolidated Financial Statements”) and in accordance with paragraph 6 of IAS 36 Impairment of assets, the CGU is the smallest identifiable group of assets corresponding to operating units that generate cash inflows that are largely independent of those from other assets (or group of assets). Depending on the facts and circumstances, the CGU may be:

·an integrated manufacturing facility which may also be an operating subsidiary; or
·a group of manufacturing facilities operated in concert with neither facility generating cash inflows that are largely independent from the cash inflows of the other.

In assessing the level of integration of manufacturing facilities to form a single CGU, the Company analyzes in particular the ability to substitute cash inflows from one facility with that of another through the interchangeability of products between the facilities and the centralization of various functions with transversal and integrated teams such as production planning, marketing and distribution and supply chain. In order to further support its assessment of whether cash inflows from an asset (or group of assets) are largely independent of the cash inflows from other assets (or group of assets), the Company also considers the guidance in paragraph 69 of IAS 36 and the various factors including how management monitors the entity’s operations (such as product lines, business, individual locations, districts or regional areas) or how management makes decisions about continuing or disposing of the entity’s assets and operations.

The Europe segment provides several examples where these criteria are applied given the large number of facilities and products. On the basis of fully integrated supply, production and distribution processes resulting in interchangeability of products and substitution


Securities and Exchange Commission, p.4

of cash inflows between the different plants, the Company’s flat carbon operations in Northern Europe, Southwestern Europe and Eastern Europe constitute 1 CGU while, following the acquisition of the Ilva business (subsequently renamed ArcelorMittal Italia) at the end of 2018, its flat carbon facilities in Italy constitute a separate CGU considering the ongoing discussions with the Italian government regarding an investment agreement with implications in terms of distinct business plan and business model. Long carbon operations are comprised of 3 CGUs (bars & rods, sections & merchant bars and long carbon operations in Poland ) in Europe including in each case integrated facilities producing the same type of interchangeable products and allowing therefore for substitution of cash inflows, and one CGU in Morocco structured on the basis of a different domestic customer base. In AM Downstream Solutions (“AMDS”), which is active in the distribution of long and flat products as well as value-added and customized steel solutions through further processing to meet specific customer requirements (such as construction, foundation solutions, wire solutions, tubular products for automotive, energy and mechanical applications, hot rolled and forged steel plates) and includes 86 legal entities across Europe, the limited interchangeability of products and substitution of cash inflows due to the specialized nature of the products or the geographic specificities of the customer portfolios resulted in the identification of 18 different CGUs.

Regarding the Company’s mining activities, mining operations which produce marketable ore (that could be sold to third parties to the extent the quality of the ore and required infrastructure is available to enable external sales) are defined as separate CGUs. Captive mines that exclusively or predominantly supply the steel operations do not benefit from the right ore quality to be able to sell to third parties. Therefore, captive mines are not separate CGUs and are tested for impairment together with the related steel operations since they are not able to generate cash flows largely independent of the related steel operations.

As a result of this analysis, the Company determined that it had 61 CGUs as of December 31, 2019. As described in note 5.3 to the 2019 Consolidated Financial Statements, whenever the CGUs comprising the operating segments are tested for impairment at the same time as goodwill, the CGUs are tested first, and any impairment of the assets is recorded prior to the testing of goodwill.

Determination and assessment of the reasonableness of the useful lives of PP&E


The Company maintains “asset cards” for each type of asset which include the data gathered and technical assessment, methods used to develop or maintain the useful life/estimate, any componentized element, operating capacity, impacts of technological or commercial obsolescence, changes in demand for output and any other factor influencing the useful life determination. As required by paragraph 51 of IAS 16 Property, Plant and Equipment, the Company performs an annual review of useful lives in order to identify any significant change in the expected pattern of consumption embodied in the asset with the support of the specialized knowledge of ArcelorMittal’s network of chief technical officers. This annual assessment of useful lives for assets in operations, which is formalized at the end of the financial year, is embedded in the overall monitoring process of the Company’s annual capital expenditure which is a critical factor of the operating performance of ArcelorMittal’s PP&E. As part of this overall monitoring process, a continuous review is performed by ArcelorMittal’s Investment Allocation Committee (“IAC”), which includes representatives from finance, strategy, capital


Securities and Exchange Commission, p.5

goods, risk management, chief technical officers and operating subsidiaries. On a monthly basis, the IAC reviews for approval approximately 25 significant capital expenditure projects covering on an annual basis the spectrum of the Company’s key components of PP&E. All capital expenditure projects above $2 million are required to go through this process, including expenditures which may be for replacement equipment where the remaining useful life of the asset being replaced is also analyzed and the relevant asset card updated as needed. In its determination of useful lives, the IAC considers factors such as the expected use of the assets, technical or commercial obsolescence, maintenance policy and operational factors. The outcome of this review process is reconsidered as part of the formalized year-end assessment described above. In accordance with IAS 16 Property, Plant and Equipment, the Company considered this information at the level of the components significant in relation to the total cost of the item of plant and equipment. In order to ensure a consistent identification and depreciation of components across the Group in compliance with paragraphs 43-47 of IAS 16, ArcelorMittal’s granularity level for the guidance on useful lives is the component level (for example structure, shell, staves, hearth, slag granulation, large valves, blowers for a blast furnace) rather than the asset formed by the components (i.e. the blast furnace).

In addition, as described in note 5.2 to the 2019 Consolidated Financial Statements, the Company performs an in-depth review of such useful lives every five years, with the support of the chief technical officer network, which includes engineers with facility-specific expertise related to plant and equipment used in the principal production units of the Company’s operations. This level of review requires a significant amount of time and effort of local teams, chief technical officers and corporate functions and as steel manufacturing equipment has long useful lives, the benefit resulting from such an in-depth review is determined to be reasonable on a 5 year basis. The most recent review took place in 2019, following which the Company determined that no changes to the useful lives of PP&E were required. In performing this review, the Company gathered and evaluated data, including commissioning dates, designed capacities, maintenance records and programs, and asset performance history, among other attributes. Furthermore, as for the annual review, the Company considers the accumulated technical experience and knowledge sharing programs that allowed for the exchange of best maintenance practices (therefore impacting the useful life) within the chief technical officer network and the deployment of these practices across the Company’s principal production units.

Determination of groups of cash generating units for impairment testing of goodwill


As described in note 5.3 to the 2019 Consolidated Financial Statements, goodwill is tested for impairment at the level of the group of GCGUs, which corresponds to the lowest level at which goodwill is monitored for internal management purposes, i.e. the operating and reportable segment level. In performing its allocation of goodwill, the Company considered the guidance of paragraph 80 of IAS 36 Impairment of Assets and allocated goodwill to the GCGU that is expected to benefit from the synergies of the business combination. Paragraph 80 further permits aggregation of CGUs for purposes of allocation of goodwill provided that the CGU or group of CGUs to which goodwill is allocated (a) represents the lowest level within the entity at which goodwill is monitored for internal management purposes and (b) is not larger than an operating segment as defined by paragraph 5 of IFRS 8, before aggregation. The Company informs the DCFOM that segment managers as defined in paragraph 9 of IFRS 8 Operating segments receive financial performance information at a CGU level below the operating segment


Securities and Exchange Commission, p.6

level or GCGU level. However, no monitoring of goodwill happens below the operating segment level as none of the operating unit managers have goodwill among their key performance indicators; and goodwill is not included in their assessment of return on capital employed. Goodwill is only included in the carrying amount of capital employed for the purpose of financial performance assessment at the operating segment level. Accordingly, as goodwill is not monitored by management at a level below the operating segment level, the Company has concluded that in accordance with IAS 36, the operating segments are the appropriate level for which goodwill should be allocated for purposes of its impairment testing and the Company includes 5 groups of cash generating units (Europe, NAFTA, Brazil, ACIS and Mining). Goodwill relating to the Mining segment was fully impaired in 2015.

Determination of the material assumptions underlying the impairment analyses and determination as to whether each material assumption is reasonable and supportable


The Company’s PP&E is dispersed among numerous geographies, is highly specialized to the steel and mining industry and often has little or no observable inputs to fair value. Except in some rare cases and excluding assets held for sale, the Company does not have readily available information on fair value of its PP&E as part of the CGU impairment assessment. Accordingly, in accordance with paragraph 20 of IAS 36 and as described in note 5.3 to the 2019 Consolidated Financial Statements, the Company generally measures the recoverability of its PP&E similarly to goodwill, i.e. based on the VIU concept.

In estimating the VIU of PP&E within the CGUs, the estimated future cash flows expected to be derived from the CGU are discounted to their present value using a post-tax discount rate (please see the section “Discount rates” below) that reflects current market assessments of the time value of money and the risks specific to the CGU. With respect to goodwill, management estimates discount rates using post-tax rates that reflect current market rates for investments of similar risk. The rate for each GCGU was estimated from the weighted average cost of capital of producers, which operate a portfolio of assets similar to those of the Company’s assets.

VIU calculations are performed combining a global corporate framework and a set of macroeconomic and business assumptions which CGUs determine based on their context of operations and facts and circumstances of markets in which they operate. The corporate framework includes the calculation of carrying amounts, tax rates and average useful lives of PP&E for the calculation of the tax shield of depreciation in the terminal value.

As described in note 5.3 to the 2019 Consolidated Financial Statements, the key assumptions for the value in use calculations are primarily the discount rates, growth rates, shipments, expected changes to average selling prices and direct costs during the period. The assumptions for the first year of the VIU calculation are aligned with the Company’s management budget by CGU or GCGU on an annual basis as approved by management and the board of directors while assumptions for year 2 to year 5 are aligned with the Company’s long-term strategy plan as approved by the appropriate management levels.



Securities and Exchange Commission, p.7



Discount rates


Discount rates were prepared by the Company with the support of an external specialist. Through the methodology described below, the Company ensures that discount rates used to calculate VIU are consistent with current market assessments using most recent market information.

To discount the projected cash flows of each CGU, the Company applied the Weighted Average Cost of Capital (“WACC”), which is the most widely used method for determining a discount rate. It implies a weighting of the cost of debt and equity at a target or optimal capital structure. The capital asset pricing model (“CAPM”) is most often used as the basis for determining the cost of equity. In order to reflect the implicit cost of capital that a potential buyer of the Company’s assets would have, the Company used trading and financial information from a sample of 39 steel and mining industry peers, taking into consideration either its industry representativeness or its similarity to ArcelorMittal’s diverse types of products (flat products, long products, mining, downstream solutions and tubular products) or geographical footprint.

In practice, the Company applies post-tax discount rates for its VIU calculations in order to align more closely with the normal business valuation approach where the use of a post-tax WACC is a generally accepted valuation principle. Using post-tax WACC is appropriate given that the WACC is market based and typically, only post-tax inputs are available to calculate the WACC (such of cost of debt). In order to comply with the IAS 36 pre-tax calculation approach, the Company applies post-tax cash flows based on notional tax cash flows. Accordingly, there are no temporary differences associated with the asset, and only future cash flows that would result if the tax base of the asset were equal to its VIU are considered. In addition, no account is taken of existing tax losses.

Paragraph 55 of IAS 36 determines that “the discount rate (rates) shall be a pre-tax rate (rates) that reflect(s) the current market assessment of: (a) the time value of money; and (b) the risks specific to the asset for which the future cash flow estimates have not been adjusted.” However, in the Basis for Conclusions of IAS 36, it is further clarified that “in theory, discounting post-tax cash flows at a post-tax discount rate and discounting pre-tax cash flows at a pre-tax discount rate should give the same result, as long as the pre-tax discount rate is the post-tax discount rate adjusted to reflect the specific amount and timing of the future tax cash flows. The pre-tax discount rate is not always the post-tax discount rate grossed up by a standard rate of tax.” (IAS 36.BCZ85). As mentioned above, the Company considers in its VIU calculations post-tax cash flows to be discounted at the WACC (by definition a post-tax rate) and, accordingly, then effectively “discounts post-tax cash flows at a post-tax rate.” The post-tax cash flows mean to apply the respective statutory tax rate to the projected earnings before interest and tax (“EBIT”) to obtain the tax expense/benefit to be considered in the calculation of the free cash flows (“FCFs”), which are then discounted at the applicable discount rates. The disclosed pre-tax WACC as required by IAS 36 is then determined by removing the tax impact from the cash flows through an iterative process with a retrofitting exercise to obtain the pre-tax WACC that equates to the same VIU.

The key factors affecting discount rates were determined as follows:


Securities and Exchange Commission, p.8



The appropriate beta factor was estimated by reference to the observed betas of comparable quoted companies in the sector, estimated based on a regression of the weekly historical return of the comparable companies against the weekly returns of the MSCI World Index over a five-year period. Observed betas were adjusted to reflect the assumption that betas converge over time toward the average of the market. The adjusted betas were unlevered based on the 5-year weighted average capital structure to assess the asset beta, which was subsequently relevered based on the debt-to-equity ratio, estimated based on the spot gearing at valuation date observed for comparable companies.

Risk-free rates


To derive USD cost of equity, the nominal risk-free rate of return was estimated by reference to the three-month average USD 30-year yield curve, which was equivalent to 2.30%.

Market risk premium


The market risk premium was estimated to be 5.50% based on published articles, academic studies and surveys that attempt to quantify the expected equity risk premium for common stocks. Recent studies include:

·Elroy Dimson, Paul Marshand Mike Staunton (Credit Suisse Global Investment Returns Yearbook 2019), which indicates a historical Equity Risk Premium (“ERP”) of 4.3% for the US and 5.0% for Germany;
·Damodaran monthly publications from January 2013 to September 2019 for the S&P 500, which indicate a median implied ERP over that period of 5.59%; and
·Fernandez, Ortiz and Acin (2019), which concluded in a survey that over the last 10 years, the median and average ERP used by practitioners was 5.50%.


Gearing and cost of debt


The cost of debt is the cost of debt financing when, for instance, issuing a bond. Since interest expenses are deductible for tax purposes, the after-tax cost of debt is considered for the purpose of calculating cost of capital. The base rate for the cost of debt is derived from the analysis of the yield on the listed debts of ArcelorMittal and comparable bonds in the metal industry with similar rating and duration. In addition, the yield is adjusted with a maturity premium.

Country risk premium


The Company has considered a country risk premium (“CRP”) for countries that have an average credit rating below AAA. Where sufficient observable data was available, the country risk premium was estimated with the sovereign spread method by reference to historical average spread of local USD and/or EUR denominated government bonds over US Treasury


Securities and Exchange Commission, p.9

bonds and/or German government bonds, respectively. In selecting bonds in order to conduct CRP estimate using the sovereign spread method, the Company applied the following criteria:

·Government issued plain vanilla bonds, preferably with an annual or semi-annual coupon frequency;
·Issued in US dollars or EUR;
·Minimum remaining maturities of five years, preferably as close as possible to 10 years, in order to maximize liquidity while maintaining a long-term view;
·Issued at least two years ago in order to maximize the number of observable data points; and
·Maximum one-year difference between the time to maturity of the selected sovereign bond and the corresponding US Treasury bond.




In addition to the basic discount rate, the Company further assesses the specific risks related to the asset as required by paragraph BCZ20 of IAS 36 through the addition of a company specific risk premium (“CSRP”), generally with the support of a valuation specialist, which is a measure of the forecasting quality of the cash flow projections based on the specific facts and circumstances of each CGU. Examples of factors that were considered in evaluating the CSRP but not limited to are key supplier dependence, key customer dependence, key person dependence, litigation risk, management’s forecast bias or management’s forecasting quality through a retrospective assessment of budget realization.

The Company believes its calculation process for discount rates is reasonable and supportable as it is performed in accordance with the professional standards applied for valuation purposes and is based on observable market and industry peer data.

Growth rates


In order to elaborate the growth rate of the 5 year forecasts, the Company’s corporate strategy department uses the macroeconomic forecasts of Oxford Economics, a leading advisory firm for global forecasting and quantitative analysis, to obtain projections of the main macroeconomic variables (GDP, inflation (consumer price index “CPI”/PPI), interest rates and exchange rates). Oxford Economics also provides detailed forecasts of over 60 industry segments every quarter, from which the Company obtains its forecasts of the main sectors using steel output (automotive, construction, machinery, metal products, other transport equipment and electrical machinery). The Company also uses a secondary source of macroeconomic information from Capital Economics. The industry forecasts are updated every quarter, which is broadly in line with the Company’s needs and are fed into its steel demand models. The Company bases its steel consuming industry forecasts using Oxford Economics and in line with the macroeconomic outlook. In addition, for the most important industry markets, the Company uses also alternative specialist resources such as LMC Auto for automotive and IHS Markit for automotive and detailed global construction output by subsector (residential, non-residential buildings and infrastructure). The Company’s corporate strategy department uses and analyzes the historical data and external forecast for steel consuming industries in order to create short-term (quarterly) and medium to long-term (5 to 10 years) apparent steel demand forecasts.


Securities and Exchange Commission, p.10

Apparent steel consumption is the actual delivery to the market and differs from real consumption (that consumed by end-use steel consuming industries) as real consumption includes changes in inventories.

In addition, in accordance with paragraph 33 (c) and 37 of IAS 36, the Company applies a 2% perpetual growth rate for the terminal value consistent with the long-term average growth rate of the markets in which it operates, and which does not exceed the average historical growth rate over the long-term. In order to corroborate its assessment, the Company ensures consistency with the inflation rate forecasts provided by Oxford Economics.



Shipment forecasts are based on apparent steel consumption (as described above) and then built based on the Company’s historical market share in each major market. The other factors considered in the shipment forecasts are any targeted market share gains based on the long-term strategic plan of the CGU or GCGU approved by management, changes to external market environment (for example factors like imports, competitor plans, etc.) and changes to internal footprint (for example loss in shipments resulting from the temporary idling of a blast furnace due to major planned maintenance work).

Steel and raw material prices


Steel and raw material price trends are difficult to predict, particularly in the current geopolitical and economic environment. While steel and raw material (in particular iron ore and coking coal) price trends have historically been correlated, a lack of correlation or an abnormal lag in the corollary relationship between raw material and steel prices may also occur and result in a “price-cost effect”, which impacts the steel spreads, i.e. the difference between raw material prices and steel prices, a key financial performance indicator in the steel industry. Iron ore is a principal raw material used in steel production and fluctuations in its market price are typically and eventually (with the timing dependent on steel market conditions) passed through to steel prices (with any potential lags in passing on higher prices). Steel and iron ore prices are affected by supply and demand trends and inventory cycles. In terms of demand, steel and iron ore prices are sensitive to trends in cyclical industries, such as the automotive, construction, appliance, machinery, equipment and transportation industries, which are significant markets for ArcelorMittal’s products. More generally, steel and iron ore prices are sensitive to macroeconomic fluctuations in the global economy which are impacted by many factors ranging from trade and geopolitical tensions to global and regional monetary policy. In the past, substantial price decreases during periods of economic weakness have not always been offset by commensurate price increases during periods of economic strength. In addition, excess supply relative to demand for steel in local markets generally results in increased exports and drives down global prices. In terms of inventory, steel stocking and destocking cycles affect apparent demand for steel and hence steel prices and steel producers’ profitability. For example, steel distributors may accumulate substantial steel inventories in periods of low prices and, in periods of rising real demand for steel from end-users, steel distributors may sell steel from inventory (destock), thereby delaying the effective implementation of steel price increases. Conversely, steel price decreases can sometimes develop their own momentum, as customers adopt a “wait and see” attitude and destock in the expectation of further price decreases.


Securities and Exchange Commission, p.11

Steel prices


The Company’ steel price outlook segregates China and ex-China, which are both updated every year for the budgeting and strategic planning process. The medium steel ex-China pricing forecast (5-year horizon) is based on supply / demand fundamentals and the cost to supply the key markets. In regions with a net import position, the forecast is based on the combination of costs of importers factoring in total market access costs (freight, material handling, duties/tariffs etc.) and the domestic cost curve to identify the long-term equilibrium of marginal supply to derive the pricing. In regions with a net export position, the forecast is based on the expected trade flows and pricing in export markets to derive weighted average export parity pricing in the long-term.

The supply outlook is based on the Company’s database with updated views of capacity expansion projects and closures based on announcements in the industry, financial information, and historical reviews of lead time to complete projects taking into consideration the nature of the project and geopolitical risks. The demand outlook is based on the assessment of apparent steel demand as described above. The cost curve is updated based on the latest forecast of raw materials prices, foreign exchange prices, oil prices and CPI.

Compared to the medium steel ex-China pricing forecast, the medium steel China pricing forecast (5-year horizon) is also based on supply / demand fundamentals and the cost of marginal domestic producers based on the microeconomic conditions from the steel production cost curve. The supply outlook is based on the mill-by-mill capacity database with an updated view of capacity expansion projects and closures based on announcements, financial information and medium/long-term government plans. The demand outlook is based on a comprehensive bottom-up demand model covering key steel-intensive segments such as real-estate, infrastructure, machinery, automotive, home appliances, shipbuilding, etc. Each segment’s forecast is modelled according to the demographic evolution, economic development and government plans. The cost curve is based on mill-by-mill cost model and is updated with the latest structural cost changes as well as the other forecasts similar to the medium steel ex-China pricing forecast.

In order to confirm that the steel selling price assumption is reasonable and supportable, the Company ensures that appropriate factors of the steel selling price assumption have been considered by comparing it to the historical evolution of steel spreads, evolution of the year 1 to year 5 projected raw material basket (i.e., the quantity of raw materials required to produce a tonne of steel, mainly comprised of iron ore, coking coal and scrap) and correlation of such metrics to the Company’s knowledge of the steel industry, market outlook on macro-economic data for the CGU’s business and economic environment and outlook for the steel industry based on the external data abovementioned 

Raw material prices


Cost of material is the main component of direct costs as it represents approximately 70% of the Company’s total cost of sales. More specifically, iron ore, coal and scrap represent approximately 70% of the cost of material. Iron ore and coal are key inputs of the basic oxygen furnace (“BOF”) route where liquid steel is produced out of a blast furnace


Securities and Exchange Commission, p.12

whereas scrap is the main input for the electric arc furnace (“EAF”) through the use of steel already produced and recycled. Iron ore and coal prices are subject to international price setting whereas scrap prices are driven by regional markets. As regards iron ore and coking coal where international prices apply, the Company’s projections of $63 per tonne to $80 per tonne (sinter feed 62% Fe China CIF) and $145 per tonne to $170 per tonne (HCC FOB Australia), respectively, as described in note 5.3 to the 2019 Consolidated Financial Statements are based on average consensus prices prepared by a panel of analysts for the long-term projections and the assumption for the 1st year is aligned with budget assumptions. The budget assumption forecast for iron ore and coking coal pricing is based on supply / demand fundamentals and the cost positions of marginal producers on the seaborne cost curve. In order to determine the assumption is reasonable and supportable, the Company ensures that appropriate factors of the year 1 iron ore and coking coal raw material prices assumptions have been considered by comparing these to external analyst and market reports. Supply outlook is based on supply guidance of major producers and the Company’s project database based on announcements, financial information with estimations of lead time to complete projects based on historical information and considering the nature of projects and the geopolitical environment. The demand outlook is based on crude steel production forecast and iron ore/coking coal input ratios of the key raw material consuming regions. The cost curve is based on production cost of seaborne supply considering the structural cost evolution, foreign exchange effects, oil prices and CPI. In order to further support its budget estimates, the Company also compares against a basket of analysts to verify whether its projections are within a reasonable range.

As opposed to iron ore and coking coal prices, medium-term heavy melting steel (“HMS”) scrap price forecasts prepared by the Company’s corporate strategy department are based on regional markets and CGUs apply the price reference relevant for the market in which they operate (East Asia import CFR, HMS CFR Turkey, Europe FOB Rotterdam, US exports East coast delivered dock, US domestic delivered at mill and HMS East Coast FOB,…). East Asia scrap is used as a reference price for the forecasting process. East Asia scrap prices follow East Asia hot metal cost, which is a function of the prices of iron ore CFR and Australia HCC coal delivered in East Asia. The other regional scrap prices are connected via seaborne scrap trade and the differentials from East Asia scrap prices mainly reflect logistic costs. These differentials have been stable historically and are used to develop the medium-term forecast based on the forward view of the hot metal cost. In addition, the Company’s corporate strategy department has its own global scrap supply/demand model that provides a view of the scrap arising per region based on historical steel consumption and the consumption of scrap based on crude steel production by route (BOF vs EAF) factoring in the relevant scrap rates. This allows the Company to check if it can expect any fundamental changes in the scrap flows which could have an impact on the regional scrap prices. In order to determine the scrap raw materials price assumption is reasonable and supportable, the Company ensures that appropriate factors of the scrap raw material price assumption have been considered by comparing it to external analysts and market reports, including the historical correlation between scrap price, iron ore and coking coal prices.


Securities and Exchange Commission, p.13

As described above, growth rates, shipments, steel selling prices and raw material prices are derived from observable market data applied in certain cases in conjunction with the Company’s own in-house expertise. In determining the reasonability of such assumptions applied to the VIU calculation, the chief controller of each CGU considers the different abovementioned factors pertaining to each key assumption in aggregate to derive its best estimates. The reasonability of such estimates is further reviewed by the operating segment management and at the corporate level. The Company ensures the assumptions are reasonable and supportable by investigating the year-over-year variations in the value in use impairment model, variations between current and prior year value in use impairment model and comparison of forecasted assumptions to the historical trend and correlation of such assumptions to the Company’s knowledge of the steel industry, market outlook on macro-economic data for the CGU’s business and economic environment and outlook for the steel industry based on the external data abovementioned. A retrospective analysis of the past performance in terms of operating income and free cash-flow (cash flow from operations less capital expenditure) compared to initial forecast provides an additional assurance of reasonableness of forward-looking assumptions despite the context of an increasingly volatile environment. VIU may result in a value in excess of the market capitalization, but this is largely due to the key factors discussed earlier that imply that the market capitalization is short-term focused rather representative of the long-term view of VIU.

Reasonableness of fair value estimates, including reconciliation of fair value estimates to market capitalization

As described above, assumptions used for VIU calculations are largely derived from observable market data and therefore reflective of the views of market participants, although these views are not fully captured in the market capitalization of the Company for the abovementioned reasons.

IAS 36 does not require a formal reconciliation between market capitalization, fair value less cost of disposal (“FVLCD”) and VIU. Instead, the Company is required to ensure that it has made sufficient disclosures describing those factors that could result in an impairment in the next periods as required by paragraph 134(f) of IAS 36. Accordingly, note 5.3 to the 2019 Consolidated Financial Statements includes disclosures regarding the GCGUs NAFTA and ACIS, where reasonably possible changes in key assumptions on which management has based its determination of the GCGU recoverable amount could cause the carrying amount to exceed its recoverable amount. The Company disclosed:

-The amount by which the recoverable amount exceeds the carrying amount;
-The change in basis points for the pre-tax discount rate
-The % decrease in average selling prices
-The % decrease in shipments


Although the direct costs are a key assumption for the VIU calculation, the Company considers that a sensitivity related to direct costs is not meaningful in isolation given the substantial correlation between steel prices and raw material prices.


Securities and Exchange Commission, p.14

VIU is a concept that was introduced by the IASB in IAS 36 Impairment of Assets but is not a generally recognized valuation concept outside of valuations performed for IAS 36. VIU is intended to measure the value to an entity of holding and operating an asset (as opposed to fair value, which considers the exit value of the asset to market participants in a hypothetical transaction). VIU involves an assessment of the recoverability of book value that directly considers an entity’s specific facts and circumstances and may be less volatile than market prices.

In determining extent to which the recoverable amount of an asset should be based on fair value, the Company considered paragraph BCZ17 in the Basis for Conclusions of IAS 36, where “the IASC rejected the proposal that an asset’s recoverable amount should be determined by reference to its fair value (based on observable market prices or, if no observable market prices exist, estimated considering prices for similar assets and the results of discounted future cash flow calculations). The reasons are the following:

“(a) IASC believed that no preference should be given to the market’s expectation of the recoverable amount of an asset (basis for fair value when market values are available and for net selling price) over a reasonable estimate performed by the individual enterprise that owns the asset (basis for fair value when market values are not available and for value in use). For example, an enterprise may have information about future cash flows that is superior to the information available in the marketplace. Also, an enterprise may plan to use an asset in a manner different from the market’s view of the best use.

(b) market values are a way to estimate fair value but only if they reflect the fact that both parties, the acquirer and the seller, are willing to enter a transaction. If an enterprise can generate greater cash flows by using an asset than by selling it, it would be misleading to base recoverable amount on the market price of the asset because a rational enterprise would not be willing to sell the asset. Therefore, recoverable amount should not refer only to a transaction between two parties (which is unlikely to happen) but should also consider an asset’s service potential from its use by the enterprise.

(c) IASC believed that in assessing the recoverable amount of an asset, it is the amount that an enterprise can expect to recover from that asset, including the effect of synergy with other assets, that is relevant.”

In addition, in paragraph BCZ20, “the IASC believes that IAS 36 included sufficient requirements to prevent an enterprise from using assumptions different from the marketplace that are unjustified. For example, an enterprise is required to determine value in use using the below information as described by the Company in its above answers:

(a) cash flow projections based on reasonable and supportable assumptions and giving greater weight to external evidence; and

(b) a discount rate that reflects current market assessments of the time value of money and the risks specific to the asset.”



Securities and Exchange Commission, p.15

As the Company utilized VIU rather than fair value in measuring the recoverability of its PP&E and its goodwill, fair value estimates were not applicable in the current year and thus a quantitative reconciliation to market capitalization was not appropriate.


* * * * *


Please advise us if we can provide any further information or assistance to facilitate your review. Please direct any further comments or questions regarding this response letter to the undersigned (+33-1-4074-6924) or Monica Kays (+33-1-4074-6983).




/s/ John D. Brinitzer

 John D. Brinitzer


cc:Anne Van Ysendyck, Esq.