XML 22 R11.htm IDEA: XBRL DOCUMENT

Allowance for Credit Losses

9 Months Ended

Sep. 30, 2020

Allowance for Credit Losses Methodology

The Allowance for Credit Losses is an important element in the Bank's financial statements. In accordance with CECL, the ACL represents management's estimate of lifetime credit losses for assets within its scope, specifically loans and leases and unfunded commitments. To calculate the ACL, management uses models to estimate the PD and LGD for loans utilizing inputs that include forecasted future economic conditions and that are dependent upon specific macroeconomic variables relevant to each of the Bank's loan and lease portfolios. Moody's Analytics, a third party, provided the historical and forward-looking macroeconomic data used in the development of the models used to calculate the ACL.

For ACL calculation purposes, the Bank considered the financial and economic environment at the time of assessment and economic scenarios that differed in the levels of severity and sensitivity to the ACL results. At each measurement date, the Bank selects the scenario that reflects its view of future economic conditions and is determined to be the most probable outcome.

All forecasts are updated for each variable where applicable and incorporated as relevant into the ACL calculation. Actual credit loss results will differ from the estimate of credit losses, either in a strong economy or a recession, as our portfolio will change through time due to growth, risk mitigation actions and other factors. In addition, the scenarios used will differ and change through time as economic conditions change. Economic scenarios might not capture deterioration or improvement in the economy timely enough for the Bank to be able to adequately assess the impact to the ACL.

Select macroeconomic variables are projected over the forecast period, and they could have a material impact in determining the ACL. As the length of the forecast period increases, information about the future becomes less readily available and projections are inherently less certain.

The following is a discussion of the changes in the factors that influenced management's current estimate of expected credit losses. The changes in the ACL estimate for all portfolio segments, during the three and nine months ended September 30, 2020, were primarily related to changes in the economic assumptions. The Bank opted to use Moody's Analytics August consensus economic forecast for estimating the ACL as of September 30, 2020. This scenario is based on Moody's Analytics review of a variety of surveys of baseline forecasts of the U.S. economy. These surveys vary in date of latest vintage, number of updates per year, list of variables forecast, duration of forecast, frequency of data (quarterly or annual), and the number of respondents to a survey. In the preparation of the Moody's Analytics consensus forecast, the focus is on the next three to five years, since that is the most typical duration in the surveyed results. Moody's Analytics approach is to give greater consideration to whatever forecasts were produced most recently, since they will include the most up-to-date historical information, and to those variables for which the number of surveyed responses is largest.

In the consensus scenario selected, the probability that the economy will perform better than this consensus is equal to the probability that it will perform worse and included the following factors:

•U.S. real GDP growth begins to decrease after the strong recovery but continues to be positive over both the short and long-term;

•U.S. unemployment rate average of 10.7% in the third quarter of 2020 and is expected to be 9.3% in the fourth quarter of 2020;

•very strong recovery with sustained growth in the fourth quarter of 2020, and continued slow growth thereafter;

•return to less than 5% unemployment by 2024.

The Bank uses an additional scenario that differs in terms of severity within the variables, both favorable and unfavorable, to assess sensitivity in the ACL results and to inform qualitative adjustments. The Bank selected the Moody's Analytics August S2 scenario for this analysis. In the scenario selected, there is a 75% probability that the economy will perform better, broadly speaking, and a 25% probability that it will perform worse; and the scenario includes the following factors:

•rising business bankruptcies and reduced consumer and business sentiment cause the economy to stall over the next several quarters;

•slow real GDP growth through 2021 with increases thereafter;

•U.S. unemployment rate average of 10.2% in the third quarter of 2020 and is expected to be 10.4% in the fourth quarter of 2020;

•very strong recovery in the current quarter of 2020, with substantially slower growth after the current quarter through the second quarter of 2021, then gradually increasing growth thereafter;

•return to less than 5% unemployment by 2025.

The results using the comparison scenario for sensitivity analysis were reviewed by management, but management believes the consensus scenario better reflects the estimated economic environment.

The ACL is measured on a collective (pool) basis when similar characteristics exist. The Company has selected models at the portfolio level using a risk-based approach, with larger, more complex portfolios having more complex models. Except as noted below, the macroeconomic variables that are inputs to the models are reasonable and supportable over the life of the loans in that they reasonably project the key economic variables in the near term and then converge to a long run equilibrium trend. These models produce reasonable and supportable estimates of loss over the life of the loans as the projected credit losses will also converge to a steady state in line with the variables applied. The Company measures the ACL using the following methods:

Commercial Real Estate ("CRE"): Non-owner occupied commercial real estate, multifamily, and construction loans are analyzed using a model that uses four primary property variables: Net Operating Income ("NOI"), Property Value, Property Type, and Location. For PD estimation, the model simulates potential future paths of NOI given commercial real estate market factors determined from macroeconomic forecasts. Using the resulting expected debt service coverage ratios, together with predicted loan-to-values and other variables, the model estimates PD from the range of conditional possibilities. In addition, the model estimates maturity PD capturing refinance default risk to produce a total PD for the loan. The model estimates LGD, inclusive of principal loss and liquidation expenses, empirically using predicted loan-to-value as well as certain market and other factors. The LGD calculation also includes a separate maturity risk component. The primary economic drivers in the model are GDP Growth, U.S. unemployment rate, and 10-Year Treasury yield. The model produces PD and LGD on a quarter-by-quarter basis for the life of loan.

The owner-occupied commercial real-estate portfolio utilizes a top down macroeconomic model using linear regression. This model produces portfolio level quarterly net charge-off rates for 10 years. The primary economic drivers for this model are the 7-year A vs Aa corporate bond spread and S&P 500 corporate after-tax profits.

Commercial: Non-homogeneous commercial loans and leases and residential development loans are analyzed in a multi-step process. An initial PD is estimated using a model driven by an obligor's selected financial statement ratios, together with cycle-adjusting information based on the obligor's state and industry. An initial LGD is derived separately based on collateral type using collateral value and a haircut to reflect the loss in liquidation. Another model then applies an auto-regression technique to the initial PD and LGD metrics to estimate the PD and LGD curves according to the macroeconomic scenario over a one-year reasonable and supportable forecast. The primary economic drivers in the model are the S&P 500 Stock Price Index, S&P 500 Market Volatility Index, U.S. unemployment rate, as well as appropriate yield curves and credit spreads. This model utilizes output reversion methodology, which, after one year, reverts on a straight-line basis over two years to long-term PD estimated using financial statement ratios of each obligor.

The model for the homogeneous lease and equipment finance agreement portfolio uses lease and equipment finance agreement information, such as origination and performance, as well as macroeconomic variables to calculate PD and LGD values. The PD calculation is based on survival analysis while LGD is calculated using a two-step regression. The model calculates LGD using an estimate of the probability that a defaulted lease or equipment finance agreement will have a loss, and an estimate of the loss amount. The primary economic drivers for the model are GDP, U.S. unemployment rate, and a home price growth index. The model produces PD and LGD curves at the lease level for each month in the forecast horizon.

Residential: The models for residential real estate and Home Equity Lines of Credit ("HELOC") utilize loan level variables, such as origination and performance, as well as macroeconomic variables to calculate PD and LGD. The U.S. unemployment rate and home price growth rate indexes are primary economic drivers in both the residential real estate and HELOC models. In addition, the prime rate is also a primary driver in the HELOC model. The models focus on establishing an empirical relationship between default probabilities and a set of loan-level, borrower, and macroeconomic credit risk drivers. The LGD calculation for residential real estate is based on an estimate of the probability that a defaulted loan will have a loss, and then an estimate of the loss amount. HELOCs utilize the same model using residential real estate LGD values to assign loans to cohorts based on FICO scores and loan age. The model produces PD and LGD curves at the loan level for each quarter in the forecast horizon.

Consumer: Historical net charge-off information as well as economic forecast assumptions are used to project loss rates for the Consumer segment.

All loans and leases that have not been modeled receive a loss rate via an extrapolated rate methodology. The loans and leases receiving an extrapolated rate are typically newly originated loans and leases or loans and leases without the granularity of data necessary to be modeled. Based on the vintage year, credit classification, and reporting category of the modeled loans and leases, a loss factor is calculated and applied to the non-modeled loans and leases.

Along with the quantitative factors produced by the above models, management also considers prepayment speeds and qualitative factors when determining the ACL. The Company uses a prepayment model that forecasts the constant prepayment rates based on institution specific data. Below are the nine qualitative factors considered where applicable:

•Changes in lending policies and procedures, including changes in underwriting standards and collection, charge-off, and recovery practices not considered elsewhere in estimating credit losses.

•Changes in national, regional, and local economic and business conditions and developments that affect the collectability of the portfolio, including the condition of various market segments.

•Changes in the nature and volume of the portfolio and in the terms of loans and leases.

•Changes in the experience, ability, and depth of lending management and other relevant staff.

•Changes in the volume and severity of past due loans and leases, the volume of non-accrual loans and leases, and the volume and severity of adversely classified or graded loans and leases.

•Changes in the quality of the Bank's credit review system.

•Changes in the value of the underlying collateral for collateral-dependent loans and leases.

•The existence and effect of any concentrations of credit, and changes in the level of such concentrations.

•The effect of other external factors such as competition and legal and regulatory requirements on the level of estimated credit losses in the Bank's existing portfolio.

The Company evaluated each qualitative factor as of September 30, 2020, and concluded that a $40.6 million increase to the model results was appropriate for the allowance for credit losses on loans and leases, due to model limitations in the current economic environment. The Company also determined that a $5.0 million increase to the calculated results was appropriate for the RUC related to construction loans. Models were deemed to be under predicting losses, because certain economic variables were lagging indicators and had not yet shown the effects of the current recession, and models were not appropriately capturing elevated risk associated with payment deferral programs offered by the Bank.

Loss factors from the models, prepayment speeds, and qualitative factors are input into the Company's CECL accounting application which aggregates the information. The Company then uses two methods to calculate the current expected credit loss: 1) the discounted cash flow ("DCF") method, which is used for all loans except lines of credit and 2) the non-discounted cash flow method which is used for lines of credit due to difficulty of calculating an effective interest rate when lines have yet to be drawn on. The DCF method utilizes the effective interest rate of individual assets to discount the expected credit losses adjusted for prepayments. The difference in the net present value and the amortized cost of the asset will result in the required allowance. The non-discounted cash flow method uses the exposure at default, along with the expected credit losses adjusted for prepayments to calculate the required allowance.

The following tables summarize activity related to the allowance for credit losses by portfolio segment for the three and nine months ended September 30, 2020:


	Three Months Ended September 30, 2020
(in thousands)	Commercial Real Estate		Commercial		Residential		Consumer & Other		Total
Allowance for credit losses on loans and leases
Balance, beginning of period	$	152,828	$	152,615	$	40,548	$	10,754	$	356,745

(Recapture) provision for credit losses for loans and leases (1)	(18,834)		25,603		(5,641)		657		1,785

Charge-offs	—		(15,426)		(120)		(1,100)		(16,646)
Recoveries	61		2,044		407		653		3,165
Net recoveries (charge-offs)	61		(13,382)		287		(447)		(13,481)
Balance, end of period	$	134,055	$	164,836	$	35,194	$	10,964	$	345,049
Reserve for unfunded commitments
Balance, beginning of period	$	20,808	$	2,921	$	2,061	$	578	$	26,368

(Recapture) provision for credit losses on unfunded commitments (1)	(380)		(1,198)		(542)		58		(2,062)
Balance, end of period	20,428		1,723		1,519		636		24,306
Total allowance for credit losses	$	154,483	$	166,559	$	36,713	$	11,600	$	369,355

(1) The total provision for credit losses as disclosed on the income statement includes a recapture of $61,000 for the three months ended September 30, 2020, related to an allowance for accrued interest on loans deferred due to COVID-19.


	Nine Months Ended September 30, 2020
(in thousands)	Commercial Real Estate		Commercial		Residential		Consumer & Other		Total
Allowance for credit losses on loans and leases
Balance, beginning of period	$	50,847	$	73,820	$	24,714	$	8,248	$	157,629
Impact of adoption CECL	5,077		44,009		2,099		(1,186)		49,999
Adjusted balance, beginning of period	55,924		117,829		26,813		7,062		207,628
Provision for credit losses for loans and leases (1)	77,664		96,577		7,701		6,829		188,771

Charge-offs	—		(55,583)		(274)		(4,697)		(60,554)
Recoveries	467		6,013		954		1,770		9,204
Net recoveries (charge-offs)	467		(49,570)		680		(2,927)		(51,350)
Balance, end of period	$	134,055	$	164,836	$	35,194	$	10,964	$	345,049
Reserve for unfunded commitments
Balance, beginning of period	$	534	$	2,539	$	149	$	1,884	$	5,106
Impact of adoption CECL	4,030		(487)		1,267		(1,572)		3,238
Adjusted balance, beginning of period	4,564		2,052		1,416		312		8,344
Provision (recapture) for credit losses on unfunded commitments (1)	15,864		(329)		103		324		15,962
Balance, end of period	20,428		1,723		1,519		636		24,306
Total allowance for credit losses	$	154,483	$	166,559	$	36,713	$	11,600	$	369,355

(1) The total provision for credit losses as disclosed on the income statement includes a provision of $99,000 for the nine months ended September 30, 2020, related to an allowance for accrued interest on loans deferred due to COVID-19.

The following tables summarize activity related to the allowance for loan and lease losses by loan and lease portfolio segment and the reserve for unfunded commitments for the three and nine months ended September 30, 2019:


	Three Months Ended September 30, 2019
(in thousands)	Commercial Real Estate		Commercial		Residential		Consumer & Other		Total
Allowance for credit losses on loans and leases
Balance, beginning of period	$	48,997	$	68,353	$	23,654	$	10,065	$	151,069
Provision	2,524		18,797		1,113		793		23,227
Charge-offs	(497)		(20,457)		(305)		(1,853)		(23,112)
Recoveries	177		4,263		94		570		5,104
Net charge-offs	(320)		(16,194)		(211)		(1,283)		(18,008)
Balance, end of period	$	51,201	$	70,956	$	24,556	$	9,575	$	156,288
Reserve for unfunded commitments
Balance, beginning of period	$	561	$	2,595	$	157	$	1,544	$	4,857
Provision (recapture) for credit losses on unfunded commitments	43		189		(55)		51		228
Balance, end of period	604		2,784		102		1,595		5,085
Total allowance for credit losses	$	51,805	$	73,740	$	24,658	$	11,170	$	161,373

	Nine Months Ended September 30, 2019
(in thousands)	Commercial Real Estate		Commercial		Residential		Consumer & Other		Total
Allowance for credit losses on loans and leases
Balance, beginning of period	$	47,904	$	63,957	$	22,034	$	10,976	$	144,871
Provision	5,599		46,135		2,630		1,899		56,263
Charge-offs	(3,035)		(48,364)		(507)		(5,065)		(56,971)
Recoveries	733		9,228		399		1,765		12,125
Net charge-offs	(2,302)		(39,136)		(108)		(3,300)		(44,846)
Balance, end of period	$	51,201	$	70,956	$	24,556	$	9,575	$	156,288
Reserve for unfunded commitments
Balance, beginning of period	$	628	$	2,250	$	160	$	1,485	$	4,523
(Recapture) provision for credit losses on unfunded commitments	(24)		534		(58)		110		562
Balance, end of period	604		2,784		102		1,595		5,085
Total allowance for credit losses	$	51,805	$	73,740	$	24,658	$	11,170	$	161,373

The following table presents the unfunded commitments for the period ended September 30, 2020 and 2019:


(in thousands)	Total
Unfunded loan and lease commitments
September 30, 2020	$	5,887,887
September 30, 2019	$	5,744,307

Asset Quality and Non-Performing Loans and Leases

The Bank manages asset quality and control credit risk through diversification of the loan and lease portfolio and the application of policies designed to promote sound underwriting and loan and lease monitoring practices. The Bank's Credit Quality Administration is charged with monitoring asset quality, establishing credit policies and procedures and enforcing the consistent application of these policies and procedures across the Bank. Reviews of non-performing, past due loans and leases and larger credits, designed to identify potential charges to the allowance for credit losses, and to determine the adequacy of the allowance, are conducted on an ongoing basis. These reviews consider such factors as the financial strength of borrowers, the value of the applicable collateral, loan and lease loss experience, estimated loan and lease losses, growth in the loan and lease portfolio, prevailing economic conditions and other factors.

Loans and Leases Past Due and Non-Accrual Loans and Leases

Typically, loans in a non-accrual status will not have an allowance for credit loss as they will be written down to their net realizable value or charged-off. However, the net realizable value for homogeneous leases and equipment finance agreements is determined by the LGD calculated by the CECL model and therefore leases and equipment finance agreements on non-accrual will have an allowance for credit losses until they become 181 days past due, at which time they are charged-off. The Company recognized no interest income on non-accrual loans and leases during the three and nine months ended September 30, 2020. As of September 30, 2020, loans of approximately $782.3 million are currently deferred under the CARES Act or interagency guidance and are classified as current as their contractual payments have been deferred.

The following tables present the amortized cost basis of the loans and leases past due, by loan and lease class, as of September 30, 2020 and December 31, 2019:


	September 30, 2020
(in thousands)	Greater than 30 to 59 Days Past Due		60 to 89 Days Past Due		Greater than 90 Days and Accruing		Total Past Due		Non-Accrual (1)		Current		Total Loans and Leases
Commercial real estate
Non-owner occupied term, net	$	3,069	$	8,959	$	1,154	$	13,182	$	8,030	$	3,512,564	$	3,533,776
Owner occupied term, net	3,432		414		1		3,847		4,385		2,402,866		2,411,098
Multifamily, net	1,868		—		—		1,868		—		3,387,166		3,389,034
Construction & development, net	—		—		—		—		—		757,462		757,462
Residential development, net	—		—		—		—		—		163,400		163,400
Commercial
Term, net	389		57		—		446		10,103		4,235,680		4,246,229
Lines of credit & other, net	2,960		3,080		3		6,043		143		888,596		894,782
Leases & equipment finance, net	8,454		24,057		18,063		50,574		3,764		1,442,312		1,496,650
Residential
Mortgage, net (2)	1,190		3,386		29,593		34,169		—		4,008,247		4,042,416
Home equity loans & lines, net	1,411		1,443		1,769		4,623		—		1,168,074		1,172,697
Consumer & other, net	1,511		475		346		2,332		—		316,597		318,929
Total, net of deferred fees and costs	$	24,284	$	41,871	$	50,929	$	117,084	$	26,425	$	22,282,964	$	22,426,473

(1) Loans and leases on non-accrual with an unamortized cost basis of $26.4 million had a related allowance for credit losses of $3.3 million at September 30, 2020.

(2) Includes government guaranteed GNMA mortgage loans that the Bank has the right but not the obligation to repurchase that are past due 90 days or more in relation to their original term, of which $19.3 million are classified as current as a result of COVID-19 related payment forbearance or deferral while $660,000 are classified as greater than 90 days past due.


	December 31, 2019
(in thousands)	Greater than 30 to 59 Days Past Due		60 to 89 Days Past Due		Greater than 90 Days and Accruing		Total Past Due		Non-Accrual		Current and Other (1)		Total Loans and Leases
Commercial real estate
Non-owner occupied term, net	$	—	$	—	$	121	$	121	$	2,920	$	3,542,525	$	3,545,566
Owner occupied term, net	975		470		1		1,446		4,600		2,490,042		2,496,088
Multifamily, net	—		—		—		—		—		3,514,774		3,514,774
Construction & development, net	—		—		—		—		—		678,740		678,740
Residential development, net	—		—		—		—		—		189,010		189,010
Commercial
Term, net	136		381		—		517		3,458		2,228,842		2,232,817
Lines of credit & other, net	3,548		376		36		3,960		767		1,207,666		1,212,393
Leases & equipment finance, net	10,685		11,176		3,086		24,947		14,499		1,426,043		1,465,489
Residential
Mortgage, net (2)	—		8,104		36,642		44,746		—		4,170,678		4,215,424
Home equity loans & lines, net	2,173		867		1,804		4,844		—		1,232,668		1,237,512
Consumer & other, net	2,043		948		615		3,606		—		404,265		407,871
Total, net of deferred fees and costs	$	19,560	$	22,322	$	42,305	$	84,187	$	26,244	$	21,085,253	$	21,195,684

(1) Other includes purchased credit impaired loans of $89.5 million.

(2) Includes government guaranteed GNMA mortgage loans that the Bank has the right but not the obligation to repurchase that are past due 90 days or more, totaling $4.3 million at December 31, 2019.

Collateral Dependent Loans and Leases

The following table summarizes the amortized cost basis of the collateral dependent loans and leases by the type of collateral securing the assets as of September 30, 2020. There have been no significant changes in the level of collateralization from the prior periods.


(in thousands)	Residential Real Estate		Commercial Real Estate		General Business Assets		Other		Total
Commercial real estate
Non-owner occupied term, net	$	—	$	7,786	$	—	$	—	$	7,786
Owner occupied term, net	—		3,627		—		—		3,627



Commercial
Term, net	940		59		7,963		1,227		10,189
Line of credit & other, net	—		—		143		—		143
Leases & equipment finance, net	—		—		3,764		—		3,764
Residential
Mortgage, net	207,199		—		—		—		207,199
Home equity loans & lines, net	2,359		—		—		—		2,359

Total net of deferred fees and costs	$	210,498	$	11,472	$	11,870	$	1,227	$	235,067

Troubled Debt Restructurings

At September 30, 2020 and December 31, 2019, troubled debt restructured loans of $15.8 million and $18.6 million, respectively, were classified as accruing TDR loans. The TDRs were granted in response to borrower financial difficulties, and generally provide for a temporary modification of loan repayment terms. In order for a new TDR loan to be considered for accrual status, the loan's collateral coverage generally will be greater than or equal to 100% of the loan balance, the loan is current on payments, and the borrower must either prefund an interest reserve or demonstrate the ability to make payments from a verified source of cash flow.

The following tables present TDR loans by accrual versus non-accrual status and by portfolio segment as of September 30, 2020 and December 31, 2019:


	September 30, 2020
(in thousands)	Accrual Status		Non-Accrual Status		Total Modification		# of Contracts
Commercial real estate, net	$	1,356	$	3,946	$	5,302	8
Commercial, net	1,258		7,954		9,212		2
Residential, net	13,100		—		13,100		78
Consumer & other, net	105		—		105		5
Total, net of deferred fees and costs	$	15,819	$	11,900	$	27,719	93

	December 31, 2019
(in thousands)	Accrual Status		Non-Accrual Status		Total Modification		# of Contracts
Commercial real estate, net	$	3,968	$	—	$	3,968	3
Commercial, net	4,105		—		4,105		2
Residential, net	10,460		—		10,460		54
Consumer & other, net	43		—		43		3
Total, net of deferred fees and costs	$	18,576	$	—	$	18,576	62

The following table presents loans that were determined to be TDRs during the three and nine months ended September 30, 2020 and 2019:


	Three Months Ended				Nine Months Ended
(in thousands)	September 30, 2020		September 30, 2019		September 30, 2020		September 30, 2019
Commercial real estate, net	$	—	$	—	$	—	$	118
Commercial, net	—		—		8,508		1,842
Residential, net	7,029		—		13,463		7
Consumer & other, net	—		—		74		—
Total, net of deferred fees and costs	$	7,029	$	—	$	22,045	$	1,967

For the periods presented in the table above, the outstanding recorded investment was the same pre and post modification and all modifications were combination modifications. There were $132,000 and $9.6 million in financing receivables modified as troubled debt restructurings within the previous 12 months for which there was a payment default during the three and nine months ended September 30, 2020. There were no financing receivables modified as troubled debt restructurings within the previous 12 months for which there was a payment default during the three and nine months ended September 30, 2019.

COVID-19 Related Payment Deferral and Forbearance

Due to the deterioration of the US economy resulting from the COVID-19 pandemic, the Company has had an increase in loan payment deferral and forbearance requests. Once a deferral or forbearance request is received, a late charge waiver is put in place and payments are suspended for an agreed-upon period. Accrued and unpaid interest during the deferral period will be collected upon the expiration of the deferral or on a regular repayment schedule at the end of the deferral period. For certain loan types, the maturity date may be extended one to six months to allow for full amortization. In accordance with the CARES Act and interagency guidance, these loans are generally classified based on their past due status prior to their deferral period, so they are classified as performing loans that accrue interest.

Credit Quality Indicators

Management regularly reviews loans and leases in the portfolio to assess credit quality indicators and to determine appropriate loan classification and grading. In addition, the board reviews and approves the credit quality indicators each year. The Bank differentiates its lending portfolios into homogeneous and non-homogeneous loans and leases. Homogeneous loans and leases are risk rated on a single risk rating scale based on the past due status of the loan or lease.

The Bank's risk rating methodology for its non-homogeneous loans and leases uses a dual risk rating approach to assess the credit risk. This approach uses two scales to provide a comprehensive assessment of credit default risk and recovery risk. The probability of default scale measures a borrower's credit default risk using risk ratings ranging from 1 to 16, where a higher rating represents higher risk. For non-homogeneous loans and leases, PD ratings of 1 through 9 are "pass" grades, while PD ratings of 10 and 11 are "watch" grades. PD ratings of 12-16 correspond to the regulatory-defined categories of special mention (12), substandard (13-14), doubtful (15), and loss (16). The loss given default scale measures the amount of loss that may not be recovered in the event of a default, using six alphabetic ratings from A-F, where a higher rating represents higher risk. The LGD scale quantifies recovery risk associated with an event of default and predicts the amount of loss that would be incurred on a loan or lease if a borrower were to experience a major default and includes variables that may be external to the borrower, such as industry, geographic location, and credit cycle stage. It could also include variables specific to the borrower such as their probability of default and bankruptcies as well as variables specific to the loan or lease, including collateral valuation, covenant structure and debt type. The product of the borrower's PD and a loan or lease LGD is the loan or lease expected loss, expressed as a percentage. This provides a common language of credit risk across different loans.

The PD scale estimates the likelihood that a borrower will experience a major default on any of its debt obligations within a specified time period. Examples of major defaults include payments 90 days or more past due, non-accrual classification, bankruptcy filing, or a full or partial charge-off of a loan or lease. As such, the PD scale represents the credit quality indicator for non-homogeneous loans and leases.

The credit quality indicator rating categories follow regulatory classification and can be generally described by the following groupings for loans and leases:

Pass—A pass loan or lease is a loan or lease with a credit risk level acceptable to the Bank for extending credit and maintaining normal credit monitoring.

Watch—A watch loan or lease is considered pass rated but has a heightened level of unacceptable default risk due to an emerging risk element or declining performance trend. Watch ratings are expected to be temporary, with issues resolved or manifested to the extent that a higher or lower risk rating would be appropriate within a short period of time.

Special Mention—A special mention loan or lease has potential weaknesses that deserve management's close attention. If left uncorrected, these potential weaknesses may result in deterioration of the repayment prospects for the asset or in the institution's credit position at some future date. These borrowers have an elevated probability of default but not to the point of a substandard classification.

Substandard—A substandard asset is inadequately protected by the current net worth and paying capacity of the borrower or of the collateral pledged, if any. Assets classified as substandard must have a well-defined weakness or weaknesses that jeopardize the liquidation of the debt. They are characterized by the distinct possibility that the Bank will sustain some loss if the deficiencies are not corrected.

Doubtful—Loans or leases classified as doubtful have all the weaknesses inherent in those classified as substandard with the added characteristic that the weaknesses make collection or liquidation in full, based on currently existing facts, conditions, and values, highly questionable and improbable.

Loss—Loans or leases classified as loss are considered uncollectible and of such little value that their continuance as bankable assets is not warranted.

The following table represents the amortized costs basis of the loans and leases by credit classification and vintage year by loan and lease class of financing receivable as of September 30, 2020:


(in thousands)		Term Loans Amortized Cost Basis by Origination Year												Revolving Loans Amortized Cost Basis		Revolving to Non-Revolving Loans Amortized Cost
	September 30, 2020	2020		2019		2018		2017		2016		Prior						Total
Commercial real estate:
Non-owner occupied term, net
Credit quality indicator:
	Pass	$	391,145	$	689,391	$	498,018	$	386,182	$	363,588	$	1,040,039	$	3,166	$	4,165	$	3,375,694
	Special mention	2,531		6,687		41,037		7,937		43,589		9,579		—		—		111,360
	Substandard	859		6,322		19,997		3,079		2,564		13,559		—		—		46,380
	Doubtful	—		—		—		—		—		—		—		—		—
	Loss	—		—		—		—		—		342		—		—		342
Total non-owner occupied term, net		$	394,535	$	702,400	$	559,052	$	397,198	$	409,741	$	1,063,519	$	3,166	$	4,165	$	3,533,776
Owner occupied term, net
Credit quality indicator:
	Pass	$	250,333	$	426,560	$	310,809	$	343,877	$	270,519	$	677,020	$	5,541	$	795	$	2,285,454
	Special mention	3,670		10,898		40,098		15,996		3,450		9,024		—		—		83,136
	Substandard	4,564		3,691		7,965		2,849		5,316		17,473		—		—		41,858
	Doubtful	—		—		—		—		—		220		—		—		220
	Loss	—		—		—		—		—		430		—		—		430
Total owner occupied term, net		$	258,567	$	441,149	$	358,872	$	362,722	$	279,285	$	704,167	$	5,541	$	795	$	2,411,098
Multifamily, net
Credit quality indicator:
	Pass	$	211,471	$	881,877	$	625,317	$	649,637	$	301,816	$	654,444	$	28,110	$	2,962	$	3,355,634
	Special mention	—		—		—		—		—		33,400		—		—		33,400
	Substandard	—		—		—		—		—		—		—		—		—
	Doubtful	—		—		—		—		—		—		—		—		—
	Loss	—		—		—		—		—		—		—		—		—
Total multifamily, net		$	211,471	$	881,877	$	625,317	$	649,637	$	301,816	$	687,844	$	28,110	$	2,962	$	3,389,034
Construction & development, net
Credit quality indicator:
	Pass	$	61,200	$	238,868	$	282,156	$	166,146	$	6,901	$	554	$	—	$	—	$	755,825
	Special mention	1,637		—		—		—		—		—		—		—		1,637
	Substandard	—		—		—		—		—		—		—		—		—
	Doubtful	—		—		—		—		—		—		—		—		—
	Loss	—		—		—		—		—		—		—		—		—
Total construction & development, net		$	62,837	$	238,868	$	282,156	$	166,146	$	6,901	$	554	$	—	$	—	$	757,462


(in thousands)		Term Loans Amortized Cost Basis by Origination Year												Revolving Loans Amortized Cost Basis		Revolving to Non-Revolving Loans Amortized Cost
	September 30, 2020	2020		2019		2018		2017		2016		Prior						Total
Residential development, net
Credit quality indicator:
	Pass	$	14,489	$	8,244	$	2,770	$	—	$	—	$	—	$	135,377	$	2,520	$	163,400
	Special mention	—		—		—		—		—		—		—		—		—
	Substandard	—		—		—		—		—		—		—		—		—
	Doubtful	—		—		—		—		—		—		—		—		—
	Loss	—		—		—		—		—		—		—		—		—
Total residential development, net		$	14,489	$	8,244	$	2,770	$	—	$	—	$	—	$	135,377	$	2,520	$	163,400

Total commercial real estate		$	941,899	$	2,272,538	$	1,828,167	$	1,575,703	$	997,743	$	2,456,084	$	172,194	$	10,442	$	10,254,770


Commercial:
Term, net
Credit quality indicator:
	Pass	$	2,314,701	$	352,092	$	351,231	$	254,727	$	71,041	$	235,495	$	594,893	$	28,230	$	4,202,410
	Special mention	2,985		307		1,427		786		2,993		3,596		585		435		13,114
	Substandard	995		1,038		4,441		5,038		1,414		1,320		—		15,884		30,130
	Doubtful	—		—		—		—		—		575		—		—		575
	Loss	—		—		—		—		—		—		—		—		—
Total term, net		$	2,318,681	$	353,437	$	357,099	$	260,551	$	75,448	$	240,986	$	595,478	$	44,549	$	4,246,229
Lines of credit & other, net
Credit quality indicator:
	Pass	$	17,429	$	22,326	$	25,574	$	791	$	2,558	$	594	$	734,015	$	6,303	$	809,590
	Special mention	4,037		—		—		—		77		324		38,772		4,089		47,299
	Substandard	573		517		—		—		252		940		29,218		6,390		37,890
	Doubtful	—		—		—		—		—		—		1		1		2
	Loss	—		—		—		—		—		—		1		—		1
Total lines of credit & other, net		$	22,039	$	22,843	$	25,574	$	791	$	2,887	$	1,858	$	802,007	$	16,783	$	894,782
Leases & equipment finance, net
Credit quality indicator:
	Pass	$	409,148	$	499,778	$	275,149	$	149,288	$	77,733	$	13,758	$	—	$	—	$	1,424,854
	Special mention	1,232		3,555		7,808		4,066		702		45		—		—		17,408
	Substandard	2,181		9,466		14,242		4,588		2,419		5,384		—		—		38,280
	Doubtful	892		5,194		4,809		2,937		1,205		169		—		—		15,206
	Loss	228		100		298		62		157		57		—		—		902
Total leases & equipment finance, net		$	413,681	$	518,093	$	302,306	$	160,941	$	82,216	$	19,413	$	—	$	—	$	1,496,650

Total commercial		$	2,754,401	$	894,373	$	684,979	$	422,283	$	160,551	$	262,257	$	1,397,485	$	61,332	$	6,637,661

Residential:
Mortgage, net
Credit quality indicator:
	Pass	$	580,651	$	1,281,752	$	449,023	$	462,024	$	490,798	$	744,660	$	—	$	—	$	4,008,908
	Special mention	—		—		197		1,206		142		3,030		—		—		4,575
	Substandard	—		1,405		2,911		5,287		6,951		10,488		—		—		27,042
	Doubtful	—		—		—		—		—		—		—		—		—
	Loss	—		1,269		—		190		—		432		—		—		1,891
Total mortgage, net		$	580,651	$	1,284,426	$	452,131	$	468,707	$	497,891	$	758,610	$	—	$	—	$	4,042,416


(in thousands)		Term Loans Amortized Cost Basis by Origination Year												Revolving Loans Amortized Cost Basis		Revolving to Non-Revolving Loans Amortized Cost
	September 30, 2020	2020		2019		2018		2017		2016		Prior						Total
Home equity loans & lines, net
Credit quality indicator:
	Pass	$	73	$	—	$	21	$	—	$	259	$	18,245	$	1,110,441	$	39,035	$	1,168,074
	Special mention	—		—		—		—		—		75		1,849		929		2,853
	Substandard	—		—		—		—		—		74		392		604		1,070
	Doubtful	—		—		—		—		—		—		—		—		—
	Loss	—		—		—		—		—		130		433		137		700
Total home equity loans & lines, net		$	73	$	—	$	21	$	—	$	259	$	18,524	$	1,113,115	$	40,705	$	1,172,697

Total residential		$	580,724	$	1,284,426	$	452,152	$	468,707	$	498,150	$	777,134	$	1,113,115	$	40,705	$	5,215,113

Consumer & other, net:
Credit quality indicator:
	Pass	$	19,644	$	25,713	$	13,427	$	59,830	$	28,164	$	15,626	$	152,320	$	1,874	$	316,598
	Special mention	—		49		45		233		224		122		1,225		88		1,986
	Substandard	14		33		—		40		20		9		157		59		332
	Doubtful	—		—		—		—		—		—		—		—		—
	Loss	—		—		—		—		—		11		2		—		13
Total consumer & other, net		$	19,658	$	25,795	$	13,472	$	60,103	$	28,408	$	15,768	$	153,704	$	2,021	$	318,929

Grand total		$	4,296,682	$	4,477,132	$	2,978,770	$	2,526,796	$	1,684,852	$	3,511,243	$	2,836,498	$	114,500	$	22,426,473