Reserve for Losses and Loss Adjustment Expenses

3 Months Ended

Mar. 31, 2022

Liability for Future Policy Benefits and Unpaid Claims and Claims Adjustment Expense [Abstract]

Reserve for Losses and Loss Adjustment Expenses

The following table provides a reconciliation of the beginning and ending reserve balances for losses and loss adjustment expenses (“LAE”) for the three months ended March 31:


(In thousands)	2022		2021
Reserve for losses and LAE at beginning of period	$	407,445	$	374,941
Less: Reinsurance recoverables	25,940		19,061
Net reserve for losses and LAE at beginning of period	381,505		355,880
Add provision for losses and LAE, net of reinsurance, occurring in:
Current period	24,369		47,989
Prior years	(131,227)		(15,667)
Net incurred losses and LAE during the current period	(106,858)		32,322
Deduct payments for losses and LAE, net of reinsurance, occurring in:
Current period	1		114
Prior years	909		1,872
Net loss and LAE payments during the current period	910		1,986
Net reserve for losses and LAE at end of period	273,737		386,216
Plus: Reinsurance recoverables	19,335		24,907
Reserve for losses and LAE at end of period	$	293,072	$	411,123

For the three months ended March 31, 2022, $0.9 million was paid for incurred claims and claim adjustment expenses attributable to insured events of prior years. There has been a $131.2 million favorable prior year development, including $101.2 million related to defaults notices received in April 2020 through September 2020 ("Early COVID Defaults"), during the three months ended March 31, 2022. Net reserves remaining as of March 31, 2022 for prior years are $249.4 million as a result of re-estimation of unpaid losses and loss adjustment expenses. For the three months ended March 31, 2021, $1.9 million was paid for incurred claims and claim adjustment expenses attributable to insured events of prior years. There was a $15.7 million favorable prior year development during the three months ended March 31, 2021. Net reserves remaining as of March 31, 2021 for prior years were $338.3 million as a result of re-estimation of unpaid losses and loss adjustment expenses. In both periods, the favorable prior years' loss development was the result of a re-estimation of amounts ultimately to be paid on prior year defaults in the default inventory, including the impact of previously identified defaults that cured. Original estimates are increased or decreased as additional information becomes known regarding individual claims.

Due to business restrictions, stay-at-home orders and travel restrictions implemented in March 2020 as a result of COVID-19, unemployment in the United States increased significantly in the second quarter of 2020, declining during the second half of 2020 and throughout 2021. As unemployment is one of the most common reasons for borrowers to default on their mortgage, the increase in unemployment resulted in an increase in the number of delinquencies on the mortgages that we insure and has the potential to increase claim frequencies on defaults.

In response to the COVID-19 pandemic, the United States government enacted a number of policies to provide fiscal stimulus to the economy and relief to those affected by this global disaster. Specifically, mortgage forbearance programs and foreclosure moratoriums were instituted by Federal legislation along with actions taken by the Federal Housing Finance Agency (“FHFA”), Fannie Mae and Freddie Mac (collectively the “GSEs”). The mortgage forbearance plans provide for eligible homeowners who were adversely impacted by COVID-19 to temporarily reduce or suspend their mortgage payments for up to 18 months for loans in an active COVID-19-related forbearance program as of February 28, 2021. For borrowers that have the ability to begin to pay their mortgage at the end of the forbearance period, we expect that mortgage servicers will work with them to modify their loans at which time the mortgage will be removed from delinquency status. We believe that the forbearance process could have a favorable effect on the frequency of claims that we ultimately pay.

Based on the fiscal stimulus, forbearance programs and the foreclosure moratoriums put in place and the credit characteristics of the defaulted loans, we expected the ultimate number of Early COVID Defaults that result in claims would be less than our historical default-to-claim experience. Accordingly, we recorded a reserve equal to approximately 7% of the initial

risk in force for the Early COVID Defaults. The reserve for the Early COVID Defaults had not been adjusted as of December 31, 2021.

As of March 31, 2022, the defaulted loans reported to us in the second and third quarters of 2020 have reached the end of their forbearance periods. During the first quarter of 2022, the Early COVID Defaults cured at elevated levels, and the cumulative cure rate for the Early COVID Defaults at March 31, 2022 exceeded our initial estimated cure rate implied by our 7% estimate of ultimate loss for these defaults. Based on cure activity through March 31, 2022 and our expectations for future cure activity, we lowered our estimate of ultimate loss for the Early COVID Defaults from 7% to 4% of the initial risk in force which resulted in a benefit recorded to the provision for losses of $101.2 million in the three months ended March 31, 2022. The reserve for losses and LAE at March 31, 2022 includes $136.9 million of reserves for Early COVID Defaults. It is reasonably possible that our estimate of the losses for the Early COVID Defaults could change in the near term as a result of changes in the economic environment, the continued impact of the pandemic on the economic environment, and the results of existing and future governmental programs designed to assist individuals and businesses impacted by the virus. A 100 basis point increase or decrease in our estimate of ultimate loss applied to the initial risk in force of the Early COVID Defaults would result in a corresponding increase or decrease in our reserve for losses and LAE of approximately $35 million as of March 31, 2022.

The credit characteristics of defaults reported subsequent to September 30, 2020 have trended towards those of the pre-pandemic periods and we have observed the normalization of other default patterns during this period. In addition, beginning in the fourth quarter of 2020 we observed a normalization of the proportion of unemployment claims related to permanent layoffs as compared to a higher proportion of temporary layoffs during the second and third quarters of 2020. We believe that while defaults subsequent to September 30, 2020 were impacted by the pandemic's effect on the economy, the underlying credit performance of these defaults may not be the same as the expected performance for the Early COVID Defaults that occurred following the onset of the pandemic and defaults after September 30, 2020 are more likely to transition consistent with pre-pandemic defaults. Accordingly, beginning in the fourth quarter of 2020, we resumed establishing reserves for defaults reported after September 2020 using our normal reserve methodology.