NOTE 3.     LOANS

The loan portfolio is classified based on the underlying collateral utilized to secure each loan for financial reporting purposes. This classification is consistent with the Quarterly Report of Condition and Income filed by ServisFirst Bank with the Federal Deposit Insurance Corporation (FDIC).

Commercial, financial and agricultural - Includes loans to business enterprises issued for commercial, industrial, agricultural production and/or other professional purposes. These loans are generally secured by equipment, inventory, and accounts receivable of the borrower and repayment is primarily dependent on business cash flows.

Real estate – construction – Includes loans secured by real estate to finance land development or the construction of industrial, commercial or residential buildings. Repayment is dependent upon the completion and eventual sale, refinance or operation of the related real estate project.

Owner-occupied commercial real estate mortgage – Includes loans secured by nonfarm nonresidential properties for which the primary source of repayment is the cash flow from the ongoing operations conducted by the party that owns the property.

1-4 family real estate mortgage – Includes loans secured by residential properties, including home equity lines of credit. Repayment is primarily dependent on the personal cash flow of the borrower.

Other real estate mortgage – Includes loans secured by nonowner-occupied properties, including office buildings, industrial buildings, warehouses, retail buildings, multifamily residential properties and farmland. Repayment is primarily dependent on income generated from the underlying collateral.

Consumer – Includes loans to individuals not secured by real estate. Repayment is dependent upon the personal cash flow of the borrower.

In light of the U.S. and global economic crisis brought about by the COVID-19 pandemic, the Company prioritized assisting its clients through this troubled time.  The CARES Act provides for Paycheck Protection Plan (“PPP”) loans to be made by banks to employers with less than 500 employees if they continue to employ their existing workers.  As of  December 31, 2020, the Company has funded approximately 4,900 loans for a total amount of $1.05 billion for clients under the PPP, and management expects to continue to participate in any extensions of the PPP by the Treasury Department. At  December 31, 2020, unaccreted deferred loan origination fees, net of costs, related to PPP loans totaled $17.8 million.  PPP loan origination fees recorded as an adjustment to loan yield for the year ended  December 31, 2020 were $14.1 million. PPP loans outstanding totaled $900.5 million at December 31, 2020 and are included within the Commercial, financial and agricultural loan category in the table below.  No allowance for credit losses has been recorded for PPP loans as they are fully guaranteed by the SBA.

The composition of loans at December 31, 2020 and 2019 is summarized as follows:

Changes in the allowance for credit losses during the years ended December 31, 2020, 2019 and 2018, respectively are as follows:

As described in Note 1. Summary of Significant Accounting Policies, the Company adopted ASU 2016-13 on January 1, 2020, which introduced the CECL methodology for estimating all expected losses over the life of a financial asset. Under the CECL methodology, the allowance for credit losses is measured on a collective basis for pools of loans with similar risk characteristics. For loans that do not share similar risk characteristics with the collectively evaluated pools, evaluations are performed on an individual basis. For all loan segments collectively evaluated, losses are predicted over a period of time determined to be reasonable and supportable, and at the end of the reasonable and supportable forecast period losses are reverted to long term historical averages. The estimated loan losses for all loan segments are adjusted for changes in qualitative factors not inherently considered in the quantitative analyses. At December 31, 2020, the Company utilized a reasonable and supportable forecast period of twelve months followed by a six-month straight-line reversion to long term averages.

The Company uses the discounted cash flow (“DCF”) method to estimate ACL for all loan pools except for commercial revolving lines of credit and credit cards. For all loan pools utilizing the DCF method, the Company utilizes and forecasts national unemployment rate as a loss driver. The Company also utilizes and forecasts GDP growth as a second loss driver for its agricultural and consumer loan pools. Consistent forecasts of the loss drivers are used across the loan segments. Upon implementation of CECL on January 1, 2020 and at December 31, 2020, a reasonable and supportable period of twelve months was utilized followed by a six-month straight-line reversion to long term averages. The Company leveraged economic projections from reputable and independent sources to inform its loss driver forecasts. At December 31, 2020 as compared to January 1, 2020, the Company forecasted a significantly higher national unemployment rate as well as a slightly higher national GDP growth rate. The Company expects national unemployment to remain above pre-pandemic levels over the forecast period with an improved national GDP growth rate as the economy comes back on line over the next year.

The Company uses loss rate methods to estimate expected credit losses for its commercial revolving lines of credit and credit card pools.  The commercial revolving lines of credit pool incorporates a probability of default (“PD”) and loss given default (“LGD”) modeling approach.  This approach involves estimating the pool average life and then using historical correlations of default and loss experience over time to calculate the lifetime PD and LGD.  These two inputs are then applied to the outstanding pool balance as of December 31, 2020. The credit card pool incorporates a remaining life modeling approach, which utilizes an attrition-based method to estimate the remaining life of the pool.  A quarterly average loss rate is then calculated using the Company’s historical loss data. The model reduces the pool balance quarterly on a straight-line basis over the estimated life of the pool. The quarterly loss rate is multiplied by the outstanding balance at each period-end resulting in an estimated loss for each quarter. The sum of estimated loss for all quarters is the total calculated reserve for the pool at December 31, 2020.

Each loan pool is adjusted for qualitative factors not inherently considered in the quantitative analyses. The qualitative adjustments either increase or decrease the quantitative model estimation. The Company considers factors that are relevant within the qualitative framework which include the following: lending policy, changes in nature and volume of loans, staff experience, changes in volume and trends of problem loans, concentration risk, trends in underlying collateral values, external factors, quality of loan review system and other economic conditions.

Inherent risks in the loan portfolio will differ based on type of loan. Specific risk characteristics by loan portfolio segment are listed below:

Commercial and industrial loans include risks associated with borrower’s cash flow, debt service coverage and management’s expertise. These loans are subject to the risk that the Company may have difficulty converting collateral to a liquid asset if necessary, as well as risks associated with degree of specialization, mobility and general collectability in a default situation. These commercial loans may be subject to many different types of risks, including fraud, bankruptcy, economic downturn, deteriorated or non-existent collateral, and changes in interest rates.

Real estate construction loans include risks associated with the borrower’s credit-worthiness, contractor’s qualifications, borrower and contractor performance, and the overall risk and complexity of the proposed project. Construction lending is also subject to risks associated with sub-market dynamics, including population, employment trends and household income. During times of economic stress, this type of loan has typically had a greater degree of risk than other loan types.   

Real estate mortgage loans consist of loans secured by commercial and residential real estate. Commercial real estate lending is dependent upon successful management, marketing and expense supervision necessary to maintain the property. Repayment of these loans may be adversely affected by conditions in the real estate market or the general economy. Also, commercial real estate loans typically involve relatively large loan balances to a single borrower. Residential real estate lending risks are generally less significant than those of other loans. Real estate lending risks include fluctuations in the value of real estate, bankruptcies, economic downturn and customer financial problems.

Consumer loans carry a moderate degree of risk compared to other loans. They are generally more risky than traditional residential real estate loans but less risky than commercial loans. Risk of default is usually determined by the well-being of the local economies. During times of economic stress, there is usually some level of job loss both nationally and locally, which directly affects the ability of the consumer to repay debt.

During the third quarter of 2019, the Company recorded a $7.4 million payment resulting from the termination of a Loan Guarantee Program (“LGP”) operated by the State of Alabama. The payment was recorded as an increase to the allowance specifically related to loans formerly enrolled in this program, in accordance with the Company’s established allowance review and evaluation criteria. In general, loans enrolled in the program had a collateral shortfall or other enhanced credit risk. In return for the Company’s acceptance of these higher risk loans, the Company received a guarantee of up to 50% of losses in the event of a borrower’s default.  These were loans that would have otherwise not met the Company’s loan underwriting criteria.  The program required a 1% fee on the commitment balance at origination.  As of December 31, 2020, the Company had 55 loans outstanding totaling $37.2 million that were formerly enrolled in the loan guarantee program. Of this total, $29.7 million were categorized as Pass within the Company's credit quality asset classification, $6.5 million were categorized as Special Mention and $1.0 million were categorized as Substandard. As of December 31, 2019, the Company had 71 loans outstanding totaling $42.2 million that were formerly enrolled in the loan guarantee program. Of this total, $37.0 million were categorized as Pass within the Company's credit quality asset classification, $5.2 million were categorized as Special Mention.

Changes in the allowance for credit losses, and allowance for loan losses, segregated by loan type, during the years ended December 31, 2020 and 2019, respectively, are as follows:

We maintain an allowance for credit losses on unfunded commercial lending commitments and letters of credit to provide for the risk of loss inherent in these arrangements. The allowance is computed using a methodology similar to that used to determine the allowance for credit losses for loans, modified to take into account the probability of a drawdown on the commitment.  The allowance for credit losses on unfunded loan commitments is classified as a liability account on the balance sheet within other liabilities, while the corresponding provision for these credit losses is recorded as a component of other expense.  The allowance for credit losses on unfunded commitments was $2.2 million at December 31, 2020. Prior to January 1, 2020, we calculated allowance for losses on unfunded loan commitments using an incurred losses methodology. At December 31, 2019, the allowance for unfunded commitments was $500,000.

The credit quality of the loan portfolio is summarized no less frequently than quarterly using categories similar to the standard asset classification system used by the federal banking agencies. The following table presents credit quality indicators for the loan loss portfolio segments and classes. These categories are utilized to develop the associated allowance for credit losses using historical losses adjusted for current economic conditions defined as follows:

The table below presents loan balances classified by credit quality indicator, loan type and based on year of origination as of December 31, 2020:

Loans by credit quality indicator as of December 31, 2019 were as follows:

Nonperforming loans include nonaccrual loans and loans 90 or more days past due and still accruing. Loans by performance status as of December 31, 2020 and 2019 are as follows:

Loans by past due status as of December 31, 2020 and 2019 are as follows:

There was no interest earned on nonaccrual loans for the years ended December 31, 2020 and 2019.

Loans that no longer share similar risk characteristics with the collectively evaluated pools are estimated on an individual basis. A loan is considered collateral-dependent when the borrower is experiencing financial difficulty and repayment is expected to be provided substantially through the operation or sale of the collateral. The following table summarizes collateral-dependent gross loans held for investment by collateral type as follows:

Prior to the adoption of ASU 2016-13, a loan was considered impaired when it was probable that the Company would be unable to collect all principal and interest payments due according to the contractual terms of the loan agreement. Individually identified impaired loans were measured based on the present value of expected payments using the loan’s original effective rate as the discount rate, the loan’s observable market price, or the fair value of the collateral if the loan was collateral dependent. If the recorded investment in the impaired loan exceeded the measure of fair value, a valuation allowance was established as part of the allowance for loan losses. Changes to the valuation allowance were recorded as a component of the provision for loan losses.

The following table presents details of the Company’s impaired loans as of December 31, 2019. Loans which have been fully charged off do not appear in the tables.

On March 22, 2020, the Interagency Statement was issued by banking regulators that encourages financial institutions to work prudently with borrowers who are or may be unable to meet their contractual payment obligations due to the effects of COVID-19. Additionally, Section 4013 of the CARES Act further provides that a qualified loan modification is exempt by law from classification as a TDR as defined by GAAP, from the period beginning March 1, 2020 until the earlier of December 31, 2020 or the date that is 60 days after the date on which the national emergency concerning the COVID-19 outbreak declared by the President of the United States under the National Emergencies Act terminates. The Interagency Statement was subsequently revised in April 2020 to clarify the interaction of the original guidance with Section 4013 of the CARES Act, as well as setting forth the banking regulators’ views on consumer protection considerations. In accordance with such guidance, the Bank is offering short-term modifications made in response to COVID-19 to borrowers who are current and otherwise not past due. These include short-term (180 days or less) modifications in the form of payment deferrals, fee waivers, extensions of repayment terms, or other delays in payment that are insignificant. As of December 31, 2020, there were 15 loans outstanding totaling $2.8 million that have payment deferrals in connection with the COVID-19 relief provided by the CARES Act. All of these remaining deferrals were principal and interest deferrals. The CARES Act precluded all of the ServisFirst COVID-19 loan modifications from being classified as a TDR as of December 31, 2020.

Troubled Debt Restructurings (“TDR”) at December 31, 2020 and 2019 totaled $1.5 million and $3.4 million, respectively. The following tables present loans modified in a TDR during the periods presented by portfolio segment and the financial impact of those modifications. The tables include modifications made to new TDRs, as well as renewals of existing TDRs.

The following table presents TDRs by portfolio segment which defaulted during the years ended December 31, 2020 and 2019, and which were modified in the previous twelve months (i.e., the twelve months prior to default). For purposes of this disclosure, default is defined as 90 days past due and still accruing or placement on nonaccrual status.

In the ordinary course of business, the Company has granted loans to certain related parties, including directors, and their affiliates. The interest rates on these loans were substantially the same as rates prevailing at the time of the transaction and repayment terms are customary for the type of loan. Changes in related party loans for the years ended December 31, 2020 and 2019 are as follows: