Other-Than-Temporary Impairment

12 Months Ended

Dec. 31, 2019

Other than Temporary Impairment Losses, Investments [Abstract]

Other-than-Temporary Impairment [Text Block]

Other-Than-Temporary Impairment

We evaluate our individual available-for-sale and held-to-maturity securities for other-than-temporary impairment each quarter.

Available-for-Sale Securities

We determined that none of our available-for-sale securities were other-than-temporarily impaired at December 31, 2019. At December 31, 2019, we held certain available-for-sale securities in an unrealized loss position. We consider these unrealized losses temporary because we expect to recover the entire amortized cost basis on these available-for-sale securities in an unrealized loss position and neither intend to sell these securities nor is it more likely than not that we will be required to sell these securities before the anticipated recovery of each security's remaining amortized cost basis. Additionally, there have been no shortfalls of principal or interest on any available-for-sale security.

Held-to-Maturity Securities

HFA Securities and Agency MBS. We have reviewed our investments in HFA securities and agency MBS and have determined that all unrealized losses are temporary. We do not intend to sell the investments nor is it more likely than not that we will be required to sell the investments before recovery of the amortized cost basis, and we do not consider these investments to be other-than-temporarily impaired at December 31, 2019.

Private-Label Residential MBS. To ensure consistency in determination of the other-than-temporary impairment for private-label residential MBS among all FHLBanks, the FHLBanks use an FHLBank System governance committee (the OTTI Governance Committee) and a formal process to ensure consistency in key other-than-temporary impairment modeling assumptions used for purposes of their cash-flow analyses for the majority of these securities. We use the FHLBanks' uniform framework and approved assumptions for purposes of our other-than-temporary impairment cash-flow analyses of our private-label residential MBS. We are responsible for making our own determination of impairment and the reasonableness of assumptions, inputs, and methodologies used and for performing the required present value calculations using appropriate historical cost bases and yields.

Our evaluation includes estimating the projected cash flows that we are likely to collect based on an assessment of available information, including the structure of the applicable security and certain assumptions to determine whether we will recover the entire amortized cost basis of the security, such as:

• the remaining payment terms for the security;

• prepayment speeds;

• default rates;

• loss severity on the collateral supporting each security based on underlying loan-level borrower and loan characteristics;

• expected housing price changes, with projections ranging from a decrease of 4.0 percent to an increase of 8.0 percent over the 12-month period beginning October 1, 2019. For the vast majority of markets, the projected short-term housing price changes range from an increase of 2.0 percent to an increase of 6.0 percent. Thereafter, we have projected a unique long-term forecast for each relevant geographic area based on an internally developed framework derived from historical data; and

• interest-rate assumptions.

To assess whether the entire amortized cost basis of private-label residential MBS will be recovered, cash-flow analyses for each of our private-label residential MBS were performed.

For those securities for which a credit loss was recognized during the year ended December 31, 2019, Table 8.1 presents a summary of the average projected values over the remaining lives of the securities for the significant inputs used to measure the amount of the credit loss recognized in earnings, as well as related current credit-enhancement. Credit-enhancement is defined as the percentage of subordinated tranches, over-collateralization, and other credit-enhancement, if any, in a security structure that will generally absorb losses before we will experience a credit loss on the security. The calculated averages represent the dollar-weighted average of Alt-A other-than-temporarily impaired private-label residential MBS.

Table 8.1 - Significant Inputs and Current Credit Enhancement for Securities with a Credit Loss in 2019 (dollars in thousands)
			Weighted Average of Significant Inputs						Weighted Average Current Credit Enhancement
Private-label MBS by Classification	Par Value		Projected Prepayment Rates		Projected Default Rates		Projected Loss Severities
Private-label residential MBS - Alt-A (1)	$	39,989	12.3	%	22.9	%	55.1	%	5.7	%

_______________________

(1) Securities are classified based upon the current performance characteristics of the underlying loan pool and therefore the manner in which the loan pool backing the security has been modeled (as prime, Alt-A, or subprime), rather than their classification of the security at the time of issuance.

Table 8.2 - Total MBS Other-than-Temporarily Impaired During the Life of the Security (dollars in thousands)
	December 31, 2019
Other-Than-Temporarily Impaired Investment (1)	Par Value		Amortized Cost		Carrying Value		Fair Value
Private-label residential MBS – Prime	$	3,796	$	3,474	$	2,510	$	3,590
Private-label residential MBS – Alt-A	561,825		383,439		308,368		470,192
Total other-than-temporarily impaired securities	$	565,621	$	386,913	$	310,878	$	473,782

_______________________

(1) Securities are classified based on the classifications of their underlying loan composition at the time of issuance. We instituted and are continuing litigation related to certain of the private-label MBS in which we invested. Our complaint asserts, among others, claims for untrue or misleading statements in the sale of securities. It is possible that classifications of private-label MBS as provided herein when based on classification at the time of issuance as disclosed by those securities' issuance documents, as well as other statements about the securities, are inaccurate.

Table 8.3 - Roll Forward of the Amounts Related to Credit Loss Recognized into Earnings (dollars in thousands)
	For the Year Ended December 31,
	2019			2018			2017
Balance at beginning of year	$	422,035		$	452,523		$	490,404
Additions:
Credit losses for which other-than-temporary impairment was not previously recognized	128			—			—
Additional credit losses for which an other-than-temporary impairment charge was previously recognized	1,084			532			1,454
Reductions:
Securities sold or matured during the year	(56,710		)	—			(5,600		)
Portion of increase in cash flows expected to be collected over the remaining life of the security that are recognized in the current period as interest income	(26,907		)	(31,020		)	(33,735		)
Balance at end of year	$	339,630		$	422,035		$	452,523