Implications of Prompt Corrective Action for the U.S. Credit Union Industry

Transcription

1 Implications of Prompt Corrective Action for the U.S. Credit Union Industry Clifford V. Rossi, PhD University of Maryland: College Park 6/5/2011

2 Study Executive Summary This report offers analysis and data on how current capital requirements affect credit unions ability to serve their members and recommends Congress provide the NCUA with the same regulatory flexibility to define capital as their bank regulatory counterparts This study brings new analysis and data to understanding the issues of credit union capital requirements by focusing on the following questions and policy implications: How do current regulatory capital requirements differ between credit unions and banks and why is this important? The vast majority of credit unions, unlike commercial banks do not have access to capital in the traditional sense. And unlike commercial bank regulators, the National Credit Union Administration (NCUA) does not possess statutory authority to define capital for credit unions. Regulatory flexibility for the NCUA is essential to establishing capital definitions better reflective of the unique structure of the credit union industry. Is the conclusion of a 2004 General Accountability Office study that no changes are warranted in credit union capital supportable by empirical facts? Surprisingly, the study bases its recommendation on very little data. GAO supported its recommendation in part by concluding that the industry had not experienced capital problems and had in fact grown relative to banks. However, this assessment was based on a limited snapshot in time from the economically benign period. Looking at data from a longer window of time - including the financial crisis - paints a much different picture of the potential impact of current capital requirements on credit unions. 2 Implications of PCA for the Credit Union Industry University of Maryland: College Park

3 How are credit union capital ratios affected over time in different economic environments? A scenario analysis of credit unions explores the impact on capital ratios of different market conditions including growth periods and credit downturns. The credit union industry s ability to serve as a stable and safe harbor for deposits and as a reliable source of credit throughout economic cycles can be negatively impacted by capital requirements limiting the ability of credit unions to better serve members. What are likely responses by credit unions to changes in capital ratios? In financial simulations of credit union performance, actions taken by credit unions to improve their capital ratios can reduce credit availability, increase the cost of credit, lower member services and/or diminish access to low-risk investment vehicles. Limiting the definition of capital to net worth introduces unintended consequences that promote outcomes harming members and the economy-at-large. What does historical data on credit unions reveal in terms of understanding of how these firms manage capital requirements? NCUA data on credit unions from support the capital ratio enhancing strategies outlined in the simulation model. Well-capitalized credit unions tend to expand their lending and share-gathering activities, while less-capitalized credit unions tend to contract. Moreover, lesscapitalized credit unions tend to offer higher rates on loans and charge higher fees. The conclusions of this study are the following: Current capital requirements create perverse incentives for credit unions to improve their net worth ratios by reducing assets, loans and shares and increasing member costs. Credit union historical data illustrates the existence of the PCA Trap. The GAO s study of credit union capital was based on extremely limited empirical assessment of credit union capital dynamics and from a period masking potential debilitating effects from the PCA Trap. 3 Implications of PCA for the Credit Union Industry University of Maryland: College Park

4 Current capital requirements stifle lending and economic growth, limiting the ability of credit unions to serve as a countercyclical source of credit. 4 Implications of PCA for the Credit Union Industry University of Maryland: College Park

5 Table of Contents Executive Summary 2 Section I. Credit unions serve a vital role in providing financial services but are constrained by existing capital requirements 6 Section II. This report offers new analysis and data to understanding how current capital requirements affect credit unions ability to serve their members and recommends Congress provide the NCUA with the same flexibility to define capital as their bank regulatory counterparts..8 Section III. Credit unions operate under inflexible, statutorily defined capital requirements, frustrating NCUA s ability to establish capital consistent with the structural features of the industry..10 Section IV. The 2004 GAO study was limited in its scope and, as a result did not fully identify the impact of current credit union capital requirements...13 Section V. Counting only net worth to calculate PCA capital requirements can severely erode credit union capital ratios as credit unions grow or enter an economic downturn.. 18 Section VI. Credit union responses to declining capital ratios harm consumers and the overall economy, by tightening and or raising the cost of credit at the wrong time, cutting back on services and raising fees...23 Section VII. Historical credit union performance data supports the findings that PCAchallenged firms may engage in capital ratio enhancement strategies that negatively impact consumers and economic growth generally 27 Section VIII. Analysis of credit union data suggests that the PCA Trap is real, that credit unions affected by this phenomenon are forced to engage in capital ratio enhancing strategies that reduce credit and investment activity, raise consumer costs and lower service and that historical data supports these conclusions.. 39 Appendix Appendix Appendix References Implications of PCA for the Credit Union Industry University of Maryland: College Park

6 I. Credit unions serve a vital role in providing financial services, but are constrained by existing capital requirements 1 Credit unions provide essential financial services to their members in the form of a stable base of insured deposits (referred to as shares) and an array of consumer loan products and services. Perhaps at no point since the Great Depression have credit unions been of greater importance to consumers than the present financial crisis. Historically, credit unions have provided a countercyclical force by providing a ready source of lending to members and as a safe harbor for shares. However, the current regulatory framework for credit union capital introduces a set of unintended consequences that constrain lending and deposit-taking activities while increasing costs at the very time that these firms should be expanding their lending and helping contain their member s costs. Credit unions are unique from commercial banks as they operate as member-owned and controlled cooperative financial institutions. In 1998, Congress enacted the Credit Union Membership Access Act (CUMAA), which imposed Prompt Corrective Action (PCA) requirements on federally insured credit unions. PCA established capital requirements for credit unions and associated supervisory actions in the event the credit union becomes less than wellcapitalized. However, unlike other depository institutions, credit unions are not able to raise capital by issuing stock or other means. By statute, retained earnings provide the only mechanism for credit unions to raise their net worth. This definition greatly limits the ability for credit unions to manage their net worth ratios. In the years following enactment of PCA, credit unions largely have enjoyed a favorable economic environment. However, as the recent financial crisis unfolded, a general uneasiness fell over the investing public regarding the relative safety of traditionally viewed low risk investments such as bank deposit accounts and money 1 The views expressed in this study are those of the author solely and do not represent those of the Robert H. Smith School of Business or the University of Maryland. 6 Implications of PCA for the Credit Union Industry University of Maryland: College Park

7 market instruments. 2 Credit union shares offered a safe haven from the uncertainty of large and small bank failures and investment funds. Moreover, as the crisis grew, banks began to tighten underwriting terms leading to a credit crunch that has yet to abate. The economic crisis also revealed the unintended consequences of a statutorily-defined new worth ratio and the imposition of PCA regulatory requirements referred to as the PCA Trap. Without a mechanism other than retained earnings to raise capital to avoid triggering regulatory action, credit unions are forced to take actions to ensure their capital ratios remain at well capitalized levels. For example, recent evidence from the crisis indicates in fact that several well-capitalized credit unions slowed share gathering activity in light of its consequences on eroding their net worth ratios. 3 At the very time consumers are looking to move their funds into credit unions during the crisis, this further weakens credit union PCA ratios. With retained earnings the only real lever to adjust the numerator of the capital ratio, credit unions find themselves in a difficult position in reducing their asset base as a way of increasing their capital ratio. As a result, lending activity may drop and in order to reduce demand, may boost lending rates. In addition, credit unions can lower what are already relatively low rates of interest on share deposits to limit share inflows. Credit unions can also raise fees on member services, lending and share products as well as reduce noninterest operating expenses such as staffing in order to improve their retained earnings position. 2 Highly publicized bank deposit runs such as at Indy Mac Bank and Washington Mutual in 2008 illustrate the concerns on the minds or ordinary investors. Also in 2008, the Reserve Primary Fund placed a 7-day freeze on investor redemptions following a reduction in the fund s net asset value below $1; this breaking the buck phenomenon created considerable concern that it could spread to other money market funds, further exacerbating the crisis. 3 NCUA Chairman Debbie Matz, Letter to House Financial Services Committee Chairman Barney Frank, December 7, Implications of PCA for the Credit Union Industry University of Maryland: College Park

8 II. This report offers analysis and data on how current capital requirements affect credit unions ability to serve their members and recommends Congress provide the NCUA with the same regulatory flexibility to define capital as their bank regulatory counterparts This study brings new analysis and data to understanding the issues of credit union capital requirements by focusing on the following questions and policy implications: How do current regulatory capital requirements differ between credit unions and banks and why is this important? The vast majority of credit unions, unlike commercial banks do not have access to capital in the traditional sense. And unlike commercial bank regulators, the National Credit Union Administration (NCUA) does not possess statutory authority to define capital for credit unions. Regulatory flexibility for the NCUA is essential to establishing capital definitions better reflective of the unique structure of the credit union industry. Is the conclusion of a 2004 General Accountability Office study that no changes are warranted in credit union capital supportable by empirical facts? Surprisingly, the study bases its recommendation on very little data. GAO supported its recommendation in part by concluding that the industry had not experienced capital problems and had in fact grown relative to banks. However, this assessment was based on a limited snapshot in time from the economically benign period. Looking at data from a longer window of time - including the financial crisis - paints a much different picture of the potential impact of current capital requirements on credit unions. How are credit union capital ratios affected over time in different economic environments? A scenario analysis of credit unions explores the impact on capital ratios 8 Implications of PCA for the Credit Union Industry University of Maryland: College Park

9 of different market conditions including growth periods and credit downturns. The credit union industry s ability to serve as a stable and safe harbor for deposits and as a reliable source of credit throughout economic cycles can be negatively impacted by capital requirements, limiting the ability of credit unions to better serve members. What are likely responses by credit unions to changes in capital ratios? In financial simulations of credit union performance, actions taken by credit unions to improve their capital ratios can reduce credit availability, increase the cost of credit, lower member services and/or diminish access to low-risk investment vehicles. Limiting the definition of capital to net worth introduces unintended consequences that promote outcomes harming members and the economy-at-large. What does historical data on credit unions reveal in terms of understanding of how these firms manage capital requirements? NCUA data on credit unions from support the capital ratio enhancing strategies outlined in the simulation model. Well-capitalized credit unions tend to expand their lending and share-gathering activities, while less-capitalized credit unions tend to contract. Moreover, lesscapitalized credit unions tend to offer higher rates on loans and charge higher fees. The conclusions of this study are the following: Current capital requirements create perverse incentives for credit unions to improve their net worth ratios by reducing assets, loans and shares and increasing member costs. Credit union historical data illustrates the existence of the PCA Trap. The GAO s study of credit union capital was based on extremely limited empirical assessment of credit union capital dynamics and from a period masking potential debilitating effects from the PCA Trap. Current capital requirements stifle lending and economic growth, limiting the ability of credit unions to serve as a countercyclical source of credit. 9 Implications of PCA for the Credit Union Industry University of Maryland: College Park

10 Congress should enact legislation providing the NCUA the same authority as bank regulators in defining regulatory capital. III. Credit unions operate under inflexible, statutorily defined capital requirements, frustrating NCUA s ability to establish capital consistent with the structural features of the industry CUMAA was enacted in 1998, imposing PCA framework on credit unions consistent with the Federal Deposit Insurance Corporation Improvement Act (FDICIA) for banks and thrifts. By way of background, following the thrift crisis of the 1980s, policymakers recognized the need to strengthen regulatory capital standards for banks and thrift institutions. In 1991, FDICIA was introduced. FDICIA included a new set of capital requirements augmented with a set of safety and soundness standards that varied in their severity with the level of the firm s capital ratio. Importantly, Congress granted authority to bank regulators to establish capital requirements but did not do the same for the NCUA. Under PCA, five categories of capital adequacy are created defined by capital to asset ratio thresholds. 4 PCA was in part designed to guard against aggressive growth of depositories by strengthening the definitions of capital and linking capital ratios to prescribed supervisory actions of increasing severity as ratios decline. Unlike the requirements for banks and thrifts, net worth for natural person credit unions, on which the PCA requirements are based, was defined in statute. Specifically, the term net worth is defined as, the retained earnings balance of the credit union, as determined under 4 In the case of commercial banks and thrifts, a leverage ratio defined as a firm s core capital as a percentage of assets serves as the PCA capital ratio. With the introduction of the Basel capital framework, two additional riskbased capital ratios are included in the PCA determination of capital adequacy. 10 Implications of PCA for the Credit Union Industry University of Maryland: College Park

11 generally accepted accounting principles. 5 Based on the statute, the NCUA has defined a net worth ratio test for credit unions as follows: 6 1/ Credit Union New Worth/Total Assets and 2/ Net Worth = Undivided Earnings + Regular Reserves + Other Reserves + Net Income Compared to a bank s leverage ratio, the net worth ratio for natural person credit unions is far more limiting. Bank capital is bifurcated into two categories; Tier 1 and Tier 2, reflecting differences in the quality of these capital components to absorb losses. Specifically, Tier 1 capital is junior in position to take losses ahead of Tier 2 capital and thus is viewed as being of greater significance. Bank core capital (also referred to as Tier 1) includes common stockholder s equity, qualifying cumulative and noncumulative perpetual preferred stock and minority interests in consolidated subsidiaries less goodwill. Supplemental capital if applied to credit unions net worth ratios under Basel capital requirements would be considered to be Tier 2 capital. Tier 2 capital includes hybrid capital instruments, subordinated debt, perpetual preferred stock and allowance for loan and lease loss (ALLL) reserves up to 1.25% of risk-weighted assets. Clearly, for the vast majority of credit unions, the ability to raise the numerator of the PCA capital requirement is greatly limited today relative to banks. The PCA capital requirement categories based on the net worth ratio for credit unions compared to commercial banks are as shown in Figure 1. 7 Compared to banks, credit union PCA required capital levels are higher. For a credit union to be considered adequately capitalized, their net worth ratio has to be at least 6% compared to 4% for a bank. This additional requirement was to recognize the slower pace that credit unions would take to build their ratios since they cannot 5 12 USC 1790d. 6 See Appendix 1, PCA Net Worth Calculation Worksheet, Call Report Form and Instructions, NCUA, 4 th Quarter Call Report, Note that net income in some cases could be included in undivided earnings. 7 For credit unions designated as complex, the NCUA establishes a risk-based ratio. Complex credit unions are defined as firms with more than $10M in assets and the risk-based net worth ratio is greater than 6%. 11 Implications of PCA for the Credit Union Industry University of Maryland: College Park

12 issue stock. A further rationale for the additional 2% is due to the contributions credit unions make to their insurance fund (NCUSIF) and to corporate credit unions. Figure 1: Credit Union and Bank PCA Categories Credit Unions Commercial Banks Net Worth/Assets Tier 1/Assets Tier 1/RWA Total Capital/RWA Percent 10% 8% WC WC WC WC AC 7% 6% 5% 4% 3% 2% AC UC SU CU AC UC SU CU AC UC SU NA UC SU NA Definitions: WC = Well Capitalized AC = Adequately Capitalized UC = Undercapitalized SU = Significantly Undercapitalized CU = Critically Undercapitalized RWA = Risk-weighted Assets Note: Commercial bank adequately capitalized and undercapitalized PCA categories for Tier 1/Total Assets reflected above are for CAMEL 1 banks Credit unions designated as low-income and corporate credit unions have access to supplemental forms of capital in addition to retained earnings to count toward the net worth ratio. 8 While these firms possess unique features that differentiate them from natural person credit unions, it is 8 Low-income credit unions qualify for this designation if they serve members earning less than 80% of the average of all employees or less than 80% of US household median income. As of December 2009 there were 1,102 lowincome designated credit unions. 12 Implications of PCA for the Credit Union Industry University of Maryland: College Park

13 instructive to note the precedent-setting use of supplemental capital for PCA requirements in the credit union industry. For example, the NCUA allows the use of uninsured secondary capital within certain term and redemption limits, among others for low-income credit unions. Corporate credit unions which operate under a cooperative structure with ownership by other natural person credit unions provide funding and investment services to their member credit unions. 9 These firms are permitted to count paid-in-capital (PIC) and membership capital (MCA) toward their net worth calculation. Both PIC and MCA forms of capital are uninsured with PIC available to cover losses in excess of retained earnings and MCA in a more senior position to cover losses after PIC is exhausted. While the purpose of this study is not alternative forms of capital for credit unions, it is important to note that several credit unions today are able to include these forms of capital in determining their capital requirements. IV. The 2004 GAO was limited in its scope and, as a result, did not fully identify the impact of current credit union capital requirements In 2004, the GAO conducted a study of the credit union PCA requirements. Responding to challenges made by the credit union industry that PCA was hampering growth, the GAO set out to examine these issues. GAO commented on the concern that during periods of large accumulations of shares, net worth ratios could decline and subject credit unions to supervisory actions. Another issue the GAO study was responding to was the concern that PCA had the potential to reduce credit unions ability to serve their members. The GAO based its conclusions, in part, on data that did not show widespread net worth problems for credit unions between The GAO study found that, between 2000 and 9 As of December 2009, 27 firms were classified as corporate credit unions. 13 Implications of PCA for the Credit Union Industry University of Maryland: College Park

14 2003, less than 3% of credit unions had a net worth ratio below the well-capitalized level. 10 The GAO also found that credit union growth rates had during the period in question, been higher than that of banks. The GAO did not provide other empirical support; the GAO also did not consider in detail the impact of the credit union PCA requirements on lending and deposit-taking activity. During the timeframe of the study, economic conditions were generally robust. So it would not be surprising that only a small percentage of credit unions would not be well capitalized. Further, the GAO study made no rigorous empirical attempt to understand the impacts of PCA on credit union lending and deposit-taking activity. The GAO s findings contrast to the present analysis. Specifically, in order to present a more complete picture of how credit unions respond to deteriorating capital ratios, this study analyzed NCUA Call Report data from along several key indicators of credit union performance. First, consistent with the GAO study, credit union asset growth rates exceeded banks during the period that the GAO study was focused (Figure 2). However, when considering the longer period between , the trend reversed leading up to and through part of the financial crisis. Thereafter, extraordinary weakness in the banking sector drove asset growth sharply downward. Figure 2 also demonstrates the procyclical nature of asset growth for credit unions and banks. This can be seen acutely during the current financial crisis when both bank and credit union growth rates dropped at different times. Credit union asset growth rates remained somewhat more stable and in fact continued to increase until However, a notable comparison is how credit union asset growth changed during periods of economic stress from being inherently countercyclical to what we see currently. 10 GAO, Credit Unions Available Information Indicates No Compelling Need for Secondary Capital, Report to Congressional Requesters, August 2004, p Implications of PCA for the Credit Union Industry University of Maryland: College Park

15 Figure 2 Credit Union and Commercial Bank Asset Growth Rates (% annual) Credit Unions Commercial Banks Source: NCUSIF and FDIC, as of end of December reporting period. For example, during the thrift crisis of the 1980s, credit union asset growth expanded compared to banks and while it declined in the mid-1980s, remained well above bank growth rates. During the and downturns, credit unions expanded their assets at greater rates than banks. Recall that PCA requirements for credit unions were not implemented until Second, loan growth trends demonstrate that credit unions historically provide credit to their members at times when other financial institutions may be tightening credit more, another issue not fully considered in the GAO report. Figure 4 shows a pattern of loan growth generally consistent with asset growth in Figure 3. Likewise, Figure 5 shows comparative loan growth 15 Implications of PCA for the Credit Union Industry University of Maryland: College Park

16 Figure 3 Asset Growth Rates : Credit Unions and Peer Commercial Banks (%) (2.00) (4.00) (6.00) (8.00) (10.00) Credit Unions Banks Under $100M Banks $100-$300M Source: NCUA December Call Reports since Of some note is how during the current crisis, credit union lending activity declined sharply at about the same rate as for banks. The GAO contends that because credit unions maintained strong net worth ratios during a period of sharp deposit growth, the industry was not hampered by PCA requirements. However, the GAO study calls out differences between short-term capital deficiencies due to factors beyond what the credit union can control and structural and management problems that erode the capital ratio. That finding seems incongruous with the recommendation that no additional flexibility for capital definitions be provided to the industry. 16 Implications of PCA for the Credit Union Industry University of Maryland: College Park

17 Figure 4 Credit Union and Commercial Bank Loan Growth Rates (% annual) Credit Unions Commercial Banks Source: NCUSIF and FDIC, as of end of December reporting period. Figure 5 Loan Growth Rates : Credit Unions and Peer Commercial Banks (%) (2.00) (4.00) (6.00) (8.00) (10.00) (12.00) Credit Unions Banks Under $100M Banks $100-$300M Source: NCUSIF and FDIC, as of end of December reporting period. 17 Implications of PCA for the Credit Union Industry University of Maryland: College Park

18 With a number of efforts underway by bank regulators to evaluate practices and policies deemed to promote procyclical outcomes for the banking industry, revisiting PCA for credit unions seems reasonable based on the latest data. 11 Further, the GAO study completely ignores the impact of PCA on consumers. Although the vast majority of credit unions maintain net worth ratios that place them in the well-capitalized category, as of December 2010, 350 credit unions fall below that PCA designation and are subject to some form of supervisory action as described above. The GAO study fails to realize that the effects of PCA cannot be well understood by looking at the average performance of credit unions and rather this issue requires additional decomposition of the industry into wellcapitalized and less-capitalized segments. As demonstrated by this analysis, the GAO study was limited in scope and, therefore, it is difficult to extrapolate to current economic conditions, which are dramatically different than when the study was conducted. V. Counting only net worth to calculate PCA capital requirements can severely erode credit union capital ratios as credit unions grow or enter an economic downturn To illustrate the effect of the PCA Trap for credit unions, this section of the study depicts four scenarios. 12 All of the scenarios assume a credit union of $100M in total loans in the current period (Year 0) growing 5% per year. For ease of exposition total assets equal total loans. 11 William Longbrake and Clifford Rossi, Assessing the Effects of Procyclical Bank Policies and Practices, Financial Services Roundtable, Cluff Fund Study, forthcoming Detailed tables comprising each scenario are found in Appendix The results are not affected by this general assumption. 14 The 5% growth rate roughly aligns with the 10-year average growth rate between for credit unions and is consistent with an assumption for asset growth used by the NCUA in Prompt Corrective Action Reform Proposal, April 2007, p Implications of PCA for the Credit Union Industry University of Maryland: College Park

19 To maintain comparability with actual credit union experience, average income, expense, provision for loan losses and charge-offs are obtained from 2006 NCUA data. 15 Scenario 1: Stable Economic Environment Scenario 1 (Figure 6) depicts a firm in a normal economic environment. This environment is characterized by low net charge-offs, starting at $.46M each year and as a percentage of loans remains flat throughout the 5-year period. Provisions for loan losses are adjusted to the level required to maintain the allowance for loan and lease losses (ALLL) at a flat level of.67% of loans over the period. Beginning period net worth is assumed to be 7.5% of loans (assets). Of interest is the fact that in this stylized normal environment, the credit union remains wellcapitalized in all periods with both assets and net worth growing over time. Assets grow at a slightly faster rate as seen in Figure 6 than net worth; however, that growth trajectory is not so high as to cause any erosion in the regulatory net worth ratio. Establishing Scenario 1 s result as a baseline allows comparisons with other scenarios where the firm enters an economic downturn marked by relatively higher credit losses or another scenario where the firm experiences moderate growth. Scenario 2: Credit Union Growth Scenario 2 (Figure 7) depicts the credit union s lending growing at a rate of 12.5% per year. The ALLL for this scenario as well as income and expense assumptions are the same as in Scenario 1. In contrast to the previous scenario, the differences in the growth rates of net worth and assets has a clear impact on the credit union s capital ratio. With assets outpacing net worth as shown by the slope of assets steeper than that of net worth, this causes the net worth ratio to decline if no other actions are taken to balance this impact. 15 Specifically from NCUA 5300 Call Report Quarterly Aggregate Financial Performance Reports, December was used to proxy for a benign economic period. 19 Implications of PCA for the Credit Union Industry University of Maryland: College Park

20 Figure 6 Scenario 1: Stable Economic Environment Year 1 Year 2 Year 3 Year 4 Year Net Worth ($) Net Worth/Assets (%) Assets ($) Note: Left-hand axis in percent and references Net Worth/Assets while right-hand axis references Net Worth and Assets in $M. Scenario 3: Stressful Economic Environment Scenario 3 by contrast depicts a firm operating in a stressful economic environment similar to the recent crisis. In this case, the ALLL starts off at 2006 levels (.67% of loans) and increases 25% per year for years 2 through 5 based on an average of credit union growth rates from Lending growth in Scenario 3 is maintained at 5% annually as in Scenario 1 and income and expense data reflect information with the exception of provision expense. The results from Scenario 3 are shown in Figure 8. As the economy deteriorates, credit union delinquencies and provisions for loan loss increase over time, creating a drag on earnings to the point of leaving the firm in a negative net income position by Year 3 (see Appendix 2 Table A2 3). With losses rising sharply over the period, by the end of Year 4 the firm falls below the wellcapitalized threshold and would at this point be subject to an earnings retention requirement supervisory action. Without any action taken by the credit union, by the end of Year 5 the net worth ratio has declined to 6.32%. 20 Implications of PCA for the Credit Union Industry University of Maryland: College Park

21 Figure 7 Scenario 2: Moderate Loan Growth Well-Capitalized PCA Threshold Year 1 Year 2 Year 3 Year 4 Year Net Worth/Assets (%) Net Worth ($) Assets ($) Note: Left-hand axis in percent and references Net Worth/Assets while right-hand axis references Net Worth and Assets in $M. Scenario 4: Impact of Supplemental Capital Scenario 4 is identical to Scenario 3 with one exception: the inclusion of supplemental capital. In this Scenario, its is assumed that the credit union has access to supplemental capital in Year 1 (Figure 9). The form of supplemental capital present is not critical to the illustration, however, it should be acknowledged that the type of capital could pose limits to how it applies to the net worth ratio. 16 It is assumed that supplemental capital represents 1% of assets; reflecting a composite of the capital forms above. 17 The use of supplemental capital would have the immediate effect of raising the credit union s net worth ratio in Year 1, allowing it to maintain its well-capitalized 16 This issue is discussed in some detail in NCUA s Supplemental Capital White Paper prepared by the the Supplemental Capital Working Group, April 12, NCUA Supplemental Capital White Paper, p.30 highlights the estimated average increase in capital ratios for each capital type and the estimate used in this study aligns within the ranges of possible outcomes. 21 Implications of PCA for the Credit Union Industry University of Maryland: College Park

22 designation throughout the 5 year period, despite higher than normal loan losses and importantly without having to take actions impacting members. Figure 8 Scenario 3: Deteriorating Credit Environment Well-Capitalized PCA Threshold Year 1 Year 2 Year 3 Year 4 Year 5 Net Worth/Assets (%) Net Worth ($) Assets ($) Note: Left-hand axis in percent and references Net Worth/Assets while right-hand axis references Net Worth and Assets in $M. It is evident from the four simple scenarios that the PCA Trap is real. In order to preserve its net worth ratio, a credit union may be forced to lower its interest on deposit rates thereby attracting fewer deposits and resulting in fewer funds available to make loans and provide other member services. However, allowing credit unions to augment their net worth through supplemental capital could remedy this anomaly, allowing credit unions to continue providing important services to members and enhancing safety and soundness. These results also have important implications as it relates to understanding the validity of the GAO s study of credit union capital since a major conclusion supporting their recommendation not to pursue capital reforms was predicated on data from a normal economic period. 22 Implications of PCA for the Credit Union Industry University of Maryland: College Park

23 Figure 9 Scenario 4: Deteriorating Credit Environment with Supplemental Capital Well-Capitalized PCA Threshold Year 1 Year 2 Year 3 Year 4 Year Net Worth/Assets (%) Net Worth ($) Assets ($) Note: Left-hand axis in percent and references Net Worth/Assets while right-hand axis references Net Worth and Assets in $M. VI. Credit union responses to declining capital ratios harm consumers and the overall economy by tightening and or raising the cost of credit at the wrong time, cutting back on services and/or raising fees Model and Assumptions To better understand how credit unions might respond to declines in net worth ratios, a simple analytic model of their financial performance was developed. The model used a number of the assumptions featured in the previous section to maintain consistency with the baseline scenario. Additionally, the model assumed that in attempting to maximize value to its members, the credit union maximizes its ability to prudently grow its lending activity subject to meeting established capital requirements and performance (as measured by net income) targets. This 1-year horizon 23 Implications of PCA for the Credit Union Industry University of Maryland: College Park

24 model takes into account loans rates and associated fee income, and has a provision for loan losses consistent with the credit performance of its portfolio. Loans are funded by regular shares and the credit union takes into account the operating expenses needed to generate loan and share business. Details on the specification of this optimization model are found in Appendix 3. To see how credit unions might proactively address declining net worth ratios, the model considers four capital ratio enhancing strategies that could be taken by credit unions. These four strategies are: reduce deposits, reduce lending, raise fees, and reduce operating expenses. An influx of deposits to the credit union acts as a catalyst for asset generation on the balance sheet so firms experiencing deteriorating capital ratios may consider actions that lower share gathering. Lowering the rate on new deposits would be a reasonable way to reduce shares. The effect on net worth ratios from reducing deposits is both direct and indirect. Lowering share rates directly raises net income which would boost the numerator of the net worth ratio. Reducing deposits would indirectly improve the net worth ratio by reducing asset generation and thus lower the denominator of the ratio. Another way credit unions could address a capital ratio issue would be to shrink the balance sheet. This could be accomplished in a number of ways, but one method is to reduce lending activity which in turn would lower the denominator of the net worth ratio. One way to accomplish this would be to set loan rates higher than market competitors. Tightening credit standards would provide another avenue to lower lending activity, but for purposes of this analysis is not considered. Raising loan rates to reduce lending work against each other in computing net income and the effects of this would be felt in the numerator of the net worth ratio while the direct reduction in lending would also lower the denominator of the capital ratio. Two other actions incorporated in the model directly improve the numerator of the net worth ratio. One would be to raise various loan, investment and customer service fees. The other would be to reduce operating expenses. The model uses a 1-year horizon in order to simplify the 24 Implications of PCA for the Credit Union Industry University of Maryland: College Park

25 analysis and amplify the impact of actions that may otherwise be taken over several years, depending on the level of their capital ratio. In other words, the significance of the model s results are in the type of actions taken and their direction (e.g., raising loan rates) rather than the absolute level of a particular outcome. To investigate to what extent a credit union would deploy any or all of these possible strategies, the model considers two scenarios. The first scenario assumes the firm starts with a net worth ratio of 5% and is thus considered undercapitalized by PCA standards. The second scenario features a credit union that is well-capitalized with a net worth ratio of 9% at the onset of the analysis. The simulation exercise implements the model described above with a number of parameters derived from 2010 NCUA Call Report data. Key assumptions for the analysis are shown in Table 1 and reflect base assumptions drawn from the earlier analysis of capital ratios over time. In the model, the decision variables for the firm were asset rates, deposit rates, operating expenses and noninterest fees. 18 Findings The results for the less-capitalized simulation are shown in Table 2 and are compelling in their implications for credit unions facing capital ratio challenges. Credit unions forced to bring their PCA designation to well-capitalized status may undertake strategies that dampen credit availability and access to stable investments, as well as lead to higher costs and lower service levels. In order to raise the net worth ratio, lending activity declines by 9%. Facilitating that decline, the credit union raises loan rates by 30bps, or about 5.5% higher than initial period loan rates. Along with the decline in assets, deposits also fall by 10% in response to a 50bp reduction in share offer rates. Fee income rises by 25bps. The credit union reduces operating expenses as it requires fewer employees and other noninterest expenses when shrinking the balance sheet. In this scenario, all four possible responses to capital ratio deficiency were deployed raising the firm s net worth ratio from 5% to 7.01%. 18 A linear optimization model establishing an objective function that maximizes lending activity subject to a set of technical and business constraints. 25 Implications of PCA for the Credit Union Industry University of Maryland: College Park

26 Table 1 Simulation Key Assumptions and Parameters Parameters and Assumptions Value Asset Base $100 Deposit Base $86 Provision for Loan Loss/Assets 0.31% Initial Asset Rate 5.38% Initial Deposit Rate 2.66% Fees/Earning Assets 1.27% Operating Expenses/Earning Assets 3.25% Table 2 Simulation Results for Less-capitalized Credit Union Factor Simulation Result Change in Lending -9% Change in Shares -10% Change in Operating Expense % Change in Fee Income % Loan Rate Change % Share Rate Change % These results stand in stark contrast with Table 3 results for well-capitalized credit unions. In this scenario, the firm is endowed with a relatively high level of capital, (part of which could include supplemental capital) that leaves it in the well-capitalized category. Since the capital ratio is not binding on this credit union, it is able to expand lending by 20% reflecting lower rates charged on loans. Share gathering formation increases at a much more modest level of 3.25% due to a 16bp higher rate offered on new deposits. Rates on fees are lowered 13bps in this 26 Implications of PCA for the Credit Union Industry University of Maryland: College Park

27 scenario. Finally, the credit union increases operating expenses given the production-oriented relationship between employees and other noninterest costs to producing shares and assets. Table 3 Simulation Results for Well-capitalized Credit Union Factor Simulation Result Change in Lending +20% Change in Shares +3.25% Change in Operating Expense % Change in Fee Income % Loan Rate Change % Share Rate Change +0016% The disparate outcomes from these two scenarios demonstrate the unintended consequences associated with credit union PCA capital requirements. Asset and deposit growth is suppressed when firms face capital ratio issues. In these situations credit unions have little latitude than to take measures that ultimately hurt their members. Allowing the NCUA to consider other forms of capital to be included in the capital ratio calculation could provide a prudent regulatory buffer while also mitigating the negative effects from capital ratio enhancement strategies. VII. Historical credit union performance data supports the findings that PCA-challenged credit unions may be forced to engage in capital ratio enhancement strategies that negatively impact consumers and economic growth generally 27 Implications of PCA for the Credit Union Industry University of Maryland: College Park

28 Historical data further corroborate aspects of the simulation results regarding asset and deposit formation, expense control and fee generation when credit unions face PCA capital constraints. This section of the paper analyzes this data. The data analysis focuses on all credit unions reporting to the NCUA between December 2006 and December Looking back over the this period provides a glimpse into credit unions behavior directly before and after the recent economic crisis. NCUA Call Report data from these periods for individual credit unions was used for this analysis. The data was reviewed for missing information and other data anomalies; in some cases a few institutions were dropped from the final sample as a result. 19 The net worth ratios for each credit union over the period were computed and compared to the 7% well-capitalized PCA threshold to segment the population by well-capitalized and less-capitalized firms. As shown in Figure 10, the average net worth ratios of less-capitalized credit unions is significantly lower than well-capitalized credit unions. Figure 10 Net Worth Ratios: Well-capitalized and Less-capitalized Credit Unions: (%) Average Well Capitalized Less Capitalized Source: NCUA December Call Reports 19 The aggregate results from this study were compared to those reported in by the NCUA where key ratios existed. The average ratios are closely aligned to those reported by the NCUA and slight differences are due to data scrubbing efforts to eliminate spurious information by the author. 28 Implications of PCA for the Credit Union Industry University of Maryland: College Park

29 Figures 11 and 12 detail the credit performance differences between less-capitalized and wellcapitalized credit unions. Delinquency rates for less-capitalized credit unions are consistently higher than for well-capitalized firms, as is to be expected, and that the resulting weaker credit performance also translates into higher net charge-offs. With the housing market in disarray and more than half of credit union investments in real estate loans, further losses on mortgage loans would erode net worth ratios in coming years generally for the industry. 20 With limited avenues available to raise these ratios, it leaves questions as to the ability of certain credit unions to support the needs of their members as the economy recovers. The number of credit unions used in the analysis is shown in Figure 13. Not surprising, over the 5 year period of the analysis, the number of credit unions declined by 12%, in part a reflection of deteriorating economic conditions as the financial crisis grew. Another trend is the rise in the number of less-capitalized credit unions over the period. By 2009, the number of less-capitalized credit unions rose sharply to nearly 5% of all reporting institutions in the final sample (Figure 14). Figure 11 Delinquency: Well-capitalized and Less-capitalized Credit Unions: (%) Average Well Capitalized Less Capitalized Source: NCUA December Call Reports 20 NCUA 2009 Yearend Statistics for Federally Insured Credit Unions, August Implications of PCA for the Credit Union Industry University of Maryland: College Park

30 Figure 12 Net Charge-off Rates: Well-capitalized and Less-capitalized Credit Unions: (%) Average Well Capitalized Less Capitalized Source: NCUA December Call Reports Figure 13 Numbers of Well-capitalized and Less-capitalized Federal Credit Unions: (%) All Well-capitalized Undercapitalized Source: NCUA December Call Reports 30 Implications of PCA for the Credit Union Industry University of Maryland: College Park

31 Figures show average asset, average loans and average shares for well-capitalized and less-capitalized credit unions. Figures compare asset, loan and deposit growth rates of well-capitalized and less-capitalized credit unions. What emerges from Figures is a clear picture that PCA may hinder credit union size. Deposit gathering and lending dramatically shrinks for less-capitalized credit unions, while well-capitalized firms exhibit growth. In 2010 for example, lending rose less than 1% for well-capitalized firms while declining more than 8% for less-capitalized credit unions. Also note that during the period when large banks were failing and money market funds came under extreme duress, less-capitalized credit unions reduced their share base. Figure 14 Trend in Percentage of Less-capitalized Federal Credit Unions: (%) Source: NCUA December Call Reports 31 Implications of PCA for the Credit Union Industry University of Maryland: College Park

32 Figure 15 Average Assets Credit Unions: ($M) 160,000, ,000, ,000, ,000,000 80,000,000 60,000,000 40,000,000 20,000, Average Well Capitalized Undercapitalized Source: NCUA December Call Reports Figure 16 Average Loans Credit Unions: ($M) 140,000, ,000, ,000,000 80,000,000 60,000,000 40,000,000 20,000, Average Well Capitalized Undercapitalized Source: NCUA December Call Reports 32 Implications of PCA for the Credit Union Industry University of Maryland: College Park

33 Figure 17 Average Shares Credit Unions: ($M) 140,000, ,000, ,000,000 80,000,000 60,000,000 40,000,000 20,000, Average Well Capitalized Undercapitalized Source: NCUA December Call Reports Figure 18 Credit Union Asset Growth Rates: (%) (2.00) (4.00) (6.00) (8.00) (10.00) Average Well Capitalized Less Capitalized Source: NCUA December Call Reports 33 Implications of PCA for the Credit Union Industry University of Maryland: College Park

34 These statistics corroborate the simulation results showing that it would be optimal for lesscapitalized credit unions to shrink the balance sheet in order to restore their net worth ratios to healthy levels. With depositors looking for viable safe havens to place their investments, lesscapitalized credit unions were unable to accommodate those consumers. With the numbers of less-capitalized credit unions growing at a faster pace than in previous years, this shrinkage strategy is inherently procyclical, providing more evidence that public policy can impose unintended consequences on a healing economy. The model simulating credit union capital ratio strategies highlighted that raising loan and deposit rates could be an effective mechanism for controlling loan and deposit growth. And the results from the simulation exercise demonstrated that less-capitalized credit unions would optimize their performance by doing so. Figure 21 shows that average rates on regular share deposits have been extremely low in recent years. However, over this 5-year period, the share rates were consistently much lower for less-capitalized credit unions than for well-capitalized Figure 19 Credit Union Loan Growth Rates: (%) (5.00) (10.00) (15.00) Average Well Capitalized Less Capitalized Source: NCUA December Call Reports 34 Implications of PCA for the Credit Union Industry University of Maryland: College Park

35 Figure 20 Credit Union Share Growth Rates: (%) (2.00) (4.00) (6.00) Average Well Capitalized Less Capitalized Source: NCUA December Call Reports Figure 21 Credit Union Borrowing Rates: (%) Average Well Capitalized Less Capitalized Source: NCUA December Call Reports 35 Implications of PCA for the Credit Union Industry University of Maryland: College Park

36 firms. While a number of possible causes could explain this difference, this result coupled with the previous figures showing marked differences in asset, loan and share formation between well- and less-capitalized credit unions supports the PCA Trap theory. Turning to loan rates, Figures provide some evidence that less-capitalized credit unions generally post higher loan rates for 1 st lien mortgages, credit cards and new auto loans than wellcapitalized credit unions. This seems to the case during the entire study period when mortgage rates posted by less-capitalized credit unions were comparable or higher than at well-capitalized firms. Some of this could be due to differences in mix between fixed-rate and adjustable rate mortgages, however, other nonmortgage product rates during the period bear similar results directionally. A potential concern would be that less-capitalized credit unions shift their risk appetite toward loan products carrying greater margin but do not accurately adjust those returns for the incremental credit risk embedded in these loans. Providing alternative forms of capital to credit unions could mitigate the potential use of high risk lending strategies as a way of working their way out of a capital problem. Nevertheless, these historical data are consistent with the simulation model exercise results showing higher rates on assets for well-capitalized credit unions as a way to reduce assets and boost the net worth ratio. The next statistics of interest relate to fee income and operating expenses. Figure 25 shows the differences between less-capitalized and well-capitalized credit unions for fee income as a percent of total assets. Consistent with the simulation results PCA credit union categories coincide with higher ratios of fee income for less-capitalized credit unions. However, in contrast to the simulation results, less-capitalized credit unions in recent years posted higher levels of operating expense than well-capitalized firms (Figure 26). Without controlling for other factors statistically it is difficult to assess the differential impact of operational inefficiency versus cost containment as a method for raising the net worth ratio. 36 Implications of PCA for the Credit Union Industry University of Maryland: College Park

37 Figure 22 Mortgage Loan Rates for Credit Unions by Type: (%) Average Well Capitalized Less Capitalized Source: NCUA December Call Reports Figure 23 New Car Loan Rates for Credit Unions by Type: (%) Average Well Capitalized Less Capitalized Source: NCUA December Call Reports 37 Implications of PCA for the Credit Union Industry University of Maryland: College Park

38 Figure 24 Credit Card Rates for Credit Unions by Type: (%) Average Well Capitalized Less Capitalized Source: NCUA December Call Reports Figure 25 Fee Income/Total Assets for Credit Unions by Type: (%) Average Well Capitalized Less Capitalized Source: NCUA December Call Reports 38 Implications of PCA for the Credit Union Industry University of Maryland: College Park

39 Figure 26 Operating Expenses/Assets for Credit Unions by Type: (%) Average Well Capitalized Less Capitalized Source: NCUA December Call Reports VIII. Analysis of credit union data suggests that the PCA Trap is real, that credit unions affected by this phenomenon are forced to engage in capital ratio enhancing strategies that reduce credit and investment activity, raise consumer costs and lower service and that historical data supports these conclusions By statute credit unions are limited in the manner in which they can raise their PCA capital ratios as a result of their unique operating model that relies on member ownership rather than private stock issuance to support and grow the franchise. Credit unions have historically been regarded as a reliable provider of credit and savings to their member base across economic cycles. 39 Implications of PCA for the Credit Union Industry University of Maryland: College Park

40 Compared to commercial banks, credit unions have in earlier downturns provided a countercyclical lending pattern, helping to accelerate economic recovery. However, limitations on what constitutes credit union capital handicap credit union s abilities to continue to provide continuous lending and investment presence for their members. This study shows how credit unions could be subject to the PCA Trap as a result of asset growth, and exacerbated under times of economic stress. The scenario results buttress the argument that the GAO s study of credit union capital ignored the impact that varying economic conditions could have on credit union net worth ratios and their response to remediate falling ratios. This study reviewed the GAO s report on credit union capital which concluded that capital reforms were not warranted at that time. The GAO study drew its conclusions based on data from 2004 and earlier and with limited empirical assessment to support their policy recommendations. The study did not examine trends in credit union performance with respect to asset and share formation or other issues relating to potentially deleterious effects on the level of service and costs to members. In contrast, this study embarked on a comprehensive assessment of credit union performance differences between well-capitalized and less-capitalized credit unions in order to bring attention to how PCA capital standards could lead to negative consequences for credit union members. A theoretical model of credit union performance was offered, describing what strategies might optimally be deployed by credit unions facing capital ratio issues. The firm s net worth ratio was specified as one of several technical and market-related constraints. From this simple model, it was shown that a credit union when facing issues with maintaining a well-capitalized net worth ratio could choose among several options to raise the ratio. These include asset and share reductions, operating cost containment strategies and fee increases. The results from this model concluded that less-capitalized credit unions would find that they could optimally raise their net worth ratios by shrinking their balance sheet along with raising fee income levels and reducing operating expenses. 40 Implications of PCA for the Credit Union Industry University of Maryland: College Park

41 Further examination into how credit unions might address capital ratio shortfalls was conducted using NCUA credit union data. A number of interesting results from this analysis have important implications for credit union capital. It was clear that less-capitalized credit unions sharply reduced their balance sheet in the years since the financial crisis. This supports the observation that PCA has a procyclical effect on credit union lending. This was further supported by generally higher rates of interest on various loan types offered by less-capitalized credit unions. These results are consistent with the simulation outcome that showed significant reductions in lending and share growth rates. Likewise, fee income ratios at less-capitalized firms were much larger than for their well-capitalized peer group. The credit union data for the period did not show lower operating costs for less-capitalized firms, however, without further analysis controlling for differences in firm efficiency, it cannot be concluded that less-capitalized credit unions do not engage in cost containment strategies as part of a net worth restoration program. The analysis from this study concludes that policymakers should revisit the issue of credit union capital with specific focus on identifying alternatives to retained earnings such as various forms of supplemental capital that could be considered. The earlier GAO study of credit union capital is of little value as the recommendation not to modify statutory capital provisions was not supported by sufficient empirical evidence over a representative period of time. Instead, the conclusion of this report is that with sweeping reforms to the financial services regulatory framework already underway, Congress should enact changes in capital requirements that provide the NCUA flexibility to allow credit unions to access supplemental capital. This would provide the NCUA with the same regulatory tools as other financial bank regulators. Failure to undertake such a reform and to align regulatory practices across agencies risks perpetuating unintended and perverse consequences of the PCA Trap that negatively impacts credit unions, consumers, and the overall economy. 41 Implications of PCA for the Credit Union Industry University of Maryland: College Park

42 Appendix 1: PCA Worksheet 42 Implications of PCA for the Credit Union Industry University of Maryland: College Park