U.S. commercial banks have posted record profits during the past few years, with return on equity for all banks hovering around 15% and return on assets well above 1% since 1993. As a result, bank holding company stocks have been doing even better than the rest of the market, which itself has recorded sizable gains during this period.
Western Banking Quarterly is a review of banking developments in the Twelfth Federal Reserve District, and includes FRBSF’s Regional Banking Tables. It is published in the Economic Letter on the fourth Friday of January, April, July, and October.
U.S. commercial banks have posted record profits during the past few years, with return on equity for all banks hovering around 15% and return on assets well above 1% since 1993. As a result, bank holding company stocks have been doing even better than the rest of the market, which itself has recorded sizable gains during this period. At the same time, however, loan loss provisioning has been somewhat lower than usual, and the ratio of reserves for loan losses to total loans has been declining. Since this trend cannot continue indefinitely, some banking observers have been prompted to question how long the unprecedented profit growth at U.S. commercial banks can be sustained. This Economic Letter discusses the economics of loan loss provisioning, its accounting relationship to reported earnings, and its tax implications. We also examine the pattern of loan loss reserves over the past 20 years and put the recent trend in loss provisioning into perspective.
Loan loss reserves are funds that banks set aside to cover bad loans. According to the American Institute of Certified Public Accountants, the allowance for loan losses should represent an amount that, in a bank management’s judgement, approximates the current amount of loans that will not be collected. Thus, loan loss reserves should be forward-looking to absorb expected future losses. This contrasts with bank capital which, in principle, is used to cover unexpected losses.
The periodic additions to loan loss reserves, by means of loan loss provisioning, are charged against current earnings. The loan loss reserve account appears on a bank’s balance sheet as a contra asset — a deduction from the bank’s outstanding loans — to give a more accurate profile of a bank’s assets and their earnings potential. When bad loans eventually are written off, they are charged against loan loss reserves. Although, strictly speaking, loan loss provisions should be based solely on expected losses, the amount of earnings that is allocated for loss provisioning may be influenced by a number of factors.
The first factor is income-smoothing. Throughout the business cycle, the credit quality of the loan portfolio is expected to fluctuate with general economic conditions. Hence, expected loan losses tend to fall during expansions, when bank earnings are strong, and rise during recessions, when bank earnings are weak. Thus, loan loss provisioning can reinforce the cyclical pattern of a bank’s reported earnings. However, because loan loss provisions are based on management’s judgement about expected losses that have not yet materialized, the amount of provisioning is subject to some degree of managerial discretion. If the bank management wants to stabilize the bank’s earnings stream over time, it is argued that they can do so by over-provisioning when bank earnings are unusually high and by under-provisioning when bank earnings are unusually low.
The second factor is tax policy. Before the Tax Reform Act of 1986, loan loss provisions were essentially treated as tax deductible expenses. Since higher provisions reduced the bank’s tax liability, bank managers had the incentive to build up loan loss reserves by over-provisioning. In 1986, the Tax Reform Act tied the amount of tax deductible loan loss expenses to a bank’s actual charge-off experience for banks with assets over $500 million. The change in tax policy eliminates the “tax shelter” of over-provisioning for large banks. However, this also makes the tax treatment of loss provisioning backward-looking, while sound risk management requires it to be forward-looking.
The third factor is capital requirements. In the past, all of a bank’s loan loss reserves were counted towards meeting regulatory capital requirements. Under risk-based capital regulations, which were fully phased in by the end of 1992, loan loss reserves are no longer counted as a component of a bank’s primary capital, that is, Tier 1 capital, but are counted only as Tier 2 capital up to 1.25% of the bank’s weighted risk assets. Hence, from the perspective of meeting regulatory capital requirements, it is much more effective to allocate income to retained earnings, which are counted fully as Tier 1 capital, than to allocate it to loan loss reserves.
Figure 1 shows loan loss reserves, loan loss provisions, and net charge-offs, as percentages of total loans for all U.S. commercial banks from 1976 through 1996. From 1976 through 1986, provisions for loan loss systematically have been above net charge-offs, gradually propelling loan loss reserves from 1% of total loans to about 1-1/2% of outstanding loans in this ten-year period. The “over-provisioning” is consistent with tax avoidance behavior, but may also reflect the deterioration in banks’ loan portfolios. However, the absence of a noticeable cyclical pattern in loan loss reserves suggests that banks were not systematically provisioning to smooth income. In 1987, concerns about the performance of loans to less developed countries led some banks to set aside sizable loan loss provisions, and the loan loss reserves skyrocketed to over 2-1/2% of total loans and remained at an elevated level until 1992.
From 1992 onward, provisions for loan loss were very much in line with net charge-offs, suggesting that, at the aggregate level, banks simply based their provisions for loan losses on their actual charge-off experience. While this left the level of loan loss reserves almost unchanged, the ratio of loss reserves to total loans outstanding has been declining, as banks registered strong loan growth during this time period. Despite the decline in the loan loss reserve ratio, it remains high relative to the levels prevailing before its buildup in the second half of the 1980s. The ratio of reserves to delinquent loans is also at a comfortably high level. It is possible, then, that banks still are adjusting to the over-provisioning in the late 1980s. If changes in tax laws and capital regulation have lowered the level of reserves targeted by banks, the declining trend in the ratio of loan loss reserves to outstanding loans may continue further. However, this trend obviously cannot last forever. When loan loss reserves cannot be further reduced and/or economic conditions start to weaken, provisions for loan loss are expected to pick up.
Against the backdrop of record profits posted by U.S. commercial banks is the declining trend in the ratio of loan loss reserves to total loans. While somewhat disturbing, this seems to be less alarming in view of the relatively high level of loan loss reserves that banks had accumulated since the late 1980s. Changes in tax law and regulatory capital requirements have further diminished banks’ incentive to build up the loan loss reserves. As the declining trend in loan loss reserves must come to an end someday, this tail wind to bank profit growth is expected to subside in the future.
Simon Kwan
Economist
Randy O’Toole
Research Associate
Opinions expressed in FRBSF Economic Letter do not necessarily reflect the views of the management of the Federal Reserve Bank of San Francisco or of the Board of Governors of the Federal Reserve System. This publication is edited by Anita Todd and Karen Barnes. Permission to reprint portions of articles or whole articles must be obtained in writing. Please send editorial comments and requests for reprint permission to research.library@sf.frb.org