The Cost to Members of Overcapitalization

The concept of an “overcapitalization bias” in the credit union system was the topic of yesterday’s post.

One comment from financial consultant Mike Higgins showed how to calculate the cost-that is money taken from members’ pockets.  Here is his analysis:

I’d like to bring some specific concerns to the table for discussion on raising credit union capital.
The proposed regulation establishes a phased-in 10% net worth floor.  We all know that credit unions will carry a buffer to avoid going below the floor – most likely in the 1.0% to 2.0% of assets range.  So, the new net worth standard from a practical basis will be 11.0% to 12.0%.
This creates a huge safety reserve, before any consideration for CECL.  Recall that loan loss reserve is effectively net worth too, just specifically earmarked to cover anticipated loan losses.
For perspective, there are 600 credit unions with assets greater than $400 million that have net worth below 11% as of the 3/31/2021 call report.  I specifically included credit unions greater than $400 million because they will have to prepare for the higher capital standards as they approach the $500 million threshold.
The current low interest rate environment is making it difficult to accrete capital via asset growth.  This fact must be considered as part of the discussion.
Any net worth requirement is a tax on asset growth.  Raising the net worth requirement increases the growth tax.  For example, a credit union growing 10% per year with an 8% net worth target must produce a ROA (profit taken from members) of 0.80% to maintain its net worth ratio.  If the net worth requirement increases to 11%, that same credit union must now produce a 1.10% ROA.
Credit unions that cannot produce the higher ROA to support the new net worth target will have their ability to grow and serve new members stifled.  It will also make it more difficult to invest in the cooperative to maintain market relevance.
In a perverse twist, credit unions may have to take more risk because of the proposed regulation to produce the higher ROA necessary to maintain net worth.
Ultimately, a higher net worth requirement harms the people credit unions are designed to serve.
I am genuinely concerned this regulation is a gift to the competition because it erodes the advantages credit unions currently hold in the marketplace.  The concept of evaluating risk to determine necessary reserves, especially in outlier situations, is important, but I agree with Chairman Hood, it should be used as a tool and not a rule.

The Overcapitalization of the Credit Union System

Twenty-five responses were filed responding to NCUA’s request for comments on the appropriate NOL cap for the NCUSIF.  One provided an insightful context for their remarks.

This excerpt from the Ohio Credit Union League  points out a larger industry bias.  This observation is especially relevant in view of NCUA’s proposal to raise the well capitalized standard for credit unions over $500 million in assets.  This new net worth option called CCULR, would raise the well capitalized compliance standard 43% in two years, from 7% to 10%.

Here is their partial comment:

. . .we wish to register a general objection to the notion of unnecessary over-capitalization of the credit union system wherever such an idea takes root. Except for a relatively small proportion of outliers, where ordinary supervision serves as an appropriate intervention, credit unions themselves are strongly capitalized to the extent that the primary buffer (natural-person credit union capital) against shocks to individual credit unions or the credit union system, is deep and broad.

Prior to the pandemic (December 2019) the average total capital ratios for U.S. and Ohio credit unions were 11.87% and 11.89%, respectively. As the pandemic began receding (March 2021), these metrics remain thoroughly robust (10.51% and 10.53%, respectively) despite the tremendous stresses of a global pandemic, global recession, and stimulus-driven ballooning balance sheets. The abundantly healthy capital levels and ratios in credit unions served the intended purpose quite effectively and in essence, shielded NCUSIF from material impact.

The regulatory process, perhaps beneficially, engenders a bias for more capital at the credit union level (seemingly, ever-stronger balances and ever-higher ratios). Yet this bias must be tempered by business discipline to ensure that capital balances in credit unions and in the NCUSIF remain strong but not excessive, so the various costs of capital are reasonable (even supportable).

To the extent that we witness what appears to be strong NCUA bias for more capital (unnecessarily larger balances and unnecessarily higher equity ratios) and noting the nexus of this concern to NOL strategy, we draw attention to the potential disruptive and costly over-capitalization of the credit union system at the credit union level, in NCUSIF, and particularly in combination. In this context we reiterate our call for the return of the NOL to its previous strong and proven level of 1.30%


A Devastating Regulatory Burden- “The Juice is Not Worth the Squeeze”

(part 2 on NCUA’s new capital proposal)

The new capital proposal’s complexity and burden is evidenced in that eight senior staff were required at the board presentation. If the agency needed eight, how many senior credit union employees will have to be involved to respond and implement it?

The new rule is 126 pages on top of the 424 pages of the 2015 RBC rule. 530 pages in total.

This proposal imposes three different plans: the 110-year old simple leverage approach for credit unions below $500 million. For credit unions over $500 million, some will have a choice between the not implemented RBC or the new CCULR. An untold number would not qualify for CCULR under the reg, so they will be forced to use the untried RBC.

Mandating $24 Billion Addition to Restricted Reserves

The CCULR alternative raises the well-capitalized minimum to 9% on Jan 1, 2022, and 10% on Jan 1, 2024. The 117 credit unions below the 9% level would have just months to raise an estimated $3 billion to be at 9% or fall under RBC. No estimate was given for the net worth shortfall for these 117 at the 10% level.

The rule also states: Subordinated Debt would not be eligible for inclusion as capital under the CCULR framework unless the complex credit union is also a low-income designated credit union.

NCUA estimates complex credit unions eligible for CCULR must hold approximately $24 billion more as regulatory capital than the total under the RBC proposal. So why would a credit union not opt for RBC?

Regulatory Coercion

NCUA’s logic is that the “simpler” CCULR calculation will cause credit unions to opt for it since 48% of complex credit unions now have total net worth exceeding 10%.

NCUA offers this regulatory “alternative” for credit unions to avoid the burden and unknowns of RBC. All that is required is to reclassify 3% of assets now in net worth from unrestricted retained earnings to the rule-mandated legal reserves, a 43% increase.

Vice Chair Hauptman explained the tradeoff this way in his opening statement:

. . .all of that (prior wording) is an (admittedly long) way of saying that a simpler-yet-higher capital standard isn’t just useful because it saves time and effort. It’s also a way of protecting the system from the problems inherent in any risk-weighting process.

This is not choice, but regulatory blackmail. Here’s one example of how NCUA says this coercion will be applied:

While a qualifying complex credit union opting into the CCULR framework, is required to have a comprehensive written strategy for maintaining an appropriate level of capital, such strategy may be straightforward and minimally state how the credit union intends to comply with the CCULR framework, including minimum capital requirements and qualifying criteria. In contrast, complex credit unions that do not opt into the CCULR framework will be required to have a more detailed written strategy. The NCUA intends to review the written strategies during the supervisory process.

A Lack of Respect for Process and Credit Unions’ Track Record

Springing a completely new capital requirement to augment one that has yet to be tried with at best a 90-day implementation timeframe, is regulatory autocracy. It shows total disdain for credit unions’ proven record of capital management.

Throughout this rule there are many movable definitions and unexplained criteria. The definition for “complex” has no meaning other than asset size. Even that simplistic approach has moved from $50 million, then $100 million and finally to $500 million to define what “complex” means.

Using this superficial asset criteria, NCUA throws its complex blanket over every large credit union. Even the $5 billion State Farm FCU with minimal products would be included.

The proposal uses two different definitions of net worth in the numerator of the capital ratios: one for CCULR and a different one for RBC. (Page 49 proposal). Net worth means one thing and then another. This overturns GAAP accounting by whatever the regulator puts in rules.

Both capital options provide four ways to calculate average assets. This makes intra-industry comparisons at best uncertain.

Even the illusion of capital choice is incorrect. Credit unions can move from one capital plan to another within the same quarter in one part of the rule. However, in another section, NCUA can stop that. Under a new Reservation of Authority, NCUA can prevent a credit union going from one standard to the other.

Credit Unions’ Capital System Is Fundamentally Different from Banking Options

Banking regulators ended the RBC requirement in 2019. The rule requires only a simple tier 1 leverage ratio which is exactly the historically proven credit union practice for measuring capital adequacy.

NCUA has kept RBC and added another banking solution overlooking fundamental differences when comparing the two systems’ capital options and purpose.

The cooperatively funded NCUSIF can provide capital assistance if that is the best solution to address a problem.

This NCUSIF’s role supporting member owned cooperatives is very different from the FDIC’s. NCUA is authorized and has used capital injections and other support (208 guarantees) to return credit unions to self-sufficiency. The FDIC cannot assist privately owned firms to restore their financial solvency.

Banks have two primary equity sources: share capital and retained earnings. Generally, the two are split about 50/50. Bank equity comes in many forms: preferred shares, subordinated debt, public and private common equity. Moreover, there are different organizational structures: bank holding companies and stand alone to increase capital flexibility.

$1 of Credit Union Capital is Worth More than $1 of Bank Equity

Moreover, $1 of credit union retained earnings is much more valuable than $1of bank earnings. The taxation of bank earnings means that each firm must earn $1.25 to $1.50 to retain $1 in reserves. Also, shareholders expect to be paid dividends on their shares or see appreciation in their stocks’ market value.

Credit union capital is Free in all respects. All sources of bank capital have requirements that do not make them “free” when choosing options.

Action Needed Now

When Board Member Hood asked staff if the majority of credit unions opposed the 2015 final RBC rule, staff replied:

“Yes, a majority of the comment letters opposed the proposal in its entirety, and many suggested the rule be withdrawn.”

So why can this comment outcome be different? I think two factors can lead to the withdrawal of this rule:

  1. The rule lacks any data, objective criteria or historical analysis as the rationale for increasing the risk based net worth ratio. There is zero objective evidence for the rule.

When commenting, must show the history of their capital details to demonstrate the credit union’s record in managing risk. Include capital plans and other documents to demonstrate your competence.

  1. Hauptman and Hood are open to listening and learning. If they are to object to this dramatic extension of regulatory micromanagement, they must be armed with information to counter the unproven assumption that credit union capital is not sufficient.

Hood closed his Board meeting statement:

The world has changed since 2015. The reality is RBC should be a tool — not a rule. If it is effective in identifying risk, put it in the examiners’ toolbox, but the last thing the NCUA should do is impose it on credit unions as an operating model. The juice just isn’t worth the squeeze for risk-based capital because this is a regulatory burden with limited benefit. Again, we already have a risk-based net worth framework as required by law, so this is not needed.”

I would add one thought, what if there is no juice? How many credit unions faced with a 500-page new capital rule will just give up and close?

Credit unions rarely fail from too little capital. They fail because their leaders’ spirits are broken. This could be the final straw for many boards and managers.

The Most Cataclysmic NCUA Proposal Ever

Last Thursday the NCUA board asked for comment on a new capital rule that would:

  • Raise the well capitalized threshold from 7% to 10%, a 43% increase, in less than six months;
  • Establish three separate, different capital rules and provide NCUA the power to override a credit union’s choice;
  • Require credit unions that choose the new 10% option to hold $24 billion more in required regulatory net worth than under the Risk Based version passed in 2015;
  • Set 60 days only for comment with implementation of the Board’s decision by Jan. 1, 2022

If approved, the result would significantly undermine credit unions’ ability to serve their members and the system’s financial soundness.

Three Capital Standards

NCUA’s proposal introduces a whole new capital standard alongside RBC. The RBC rule from 2015 has yet to be implemented. There is no actual credit union experience with either RBC or this second higher minimum capital standard called CCULR, or Complex Credit Union Leverage Ratio.

The accelerated, short time frame for this far-reaching change is unnecessary and unwise. In the 8 years since NCUA proposed RBC, banking regulators evaluated their experience and dropped it as a requirement. Now NCUA retains RBC and adds another banking creation without any analysis or data for either option.

The most cogent reaction to this proposal was by Board Member Hood who opened his questions by saying: “Mr. Chairman, after serious study and consideration, my preference would be to table the risk-based capital rule indefinitely—or even repeal it—and fine-tune the risk-based net worth rule as needed.”

The Most Restrictive Rule Since Deregulation

This race to alter credit unions’ proven capital framework will severely lesson the system’s strength and resilience. Both options constrain the most important decisions a CEO and board make about their business model: how much and where to invest in member value versus how much to keep in reserves.

RBC is a regulatory tax on every asset transaction made by a credit union. To avoid this regulatory capital tax credit unions can use CCULR, the new “simpler” option. This would increases a credit union’s restricted earnings by 43% versus the current 7% capital requirement.

The most important operational judgments for members are no longer a credit union’s to determine. Rather, NCUA defines a single definition of balance sheet risks, weighs them, and then sets the minimum capital using an untried formula.

All credit union’s transactions are treated identically for the same asset risk whether located New York, Peoria, or Charleston. This one size fits all denies the objective reality that every credit union’s operating environment is different.

NCUA then reserves the authority to change a credit union’s decision or to modify its weightings, definitions and the required capital minimum at any time. Just as it does in the current proposal!

Existing Capital Approach Has Stabilized Credit Unions for 110 years—Risk Based Net Worth Approved by GAO

The proposition that credit union’s minimum capital standards have not been sufficient is false. NCUA provided not a single example, data or historical event to suggest otherwise.

In 1998 the well-capitalized threshold was raised from 6% to 7% as a political compromise to the suggestion that credit unions expense their 1% underwriting in the NCUSIF. There was no accounting or factual basis for this change.

In 2004, GAO reviewed NCUA’s implementation of this new PCA risk based net worth concept and concluded:

We are aware that NCUA is constructing a more detailed risk-based capital proposal . . .and that any proposal should be based on the premise that risk-based capital be used to augment, but not replace, the current net worth requirement for credit unions. The system of PCA implemented for credit unions is comparable with the PCA system that bank and thrift regulators have used for over a decade. and

. . . available information indicates no compelling need. . . to make other significant changes to PCA as it has been implemented for credit unions.

Credit unions start with no reserve capital. The cooperative approach of reserving from earnings has been sufficient to sustain the industry through both macro events and decades of financial innovation including:

  • Introducing share drafts, mortgage lending, member business loans, credit/debit cards;
  • Building out distribution capabilities including branches, call centers, Internet Retail services and shared ATM and branch networks;
  • Investing in CUSO’s to bring scale and collaborative solutions for member value;
  • Incorporating the latest financial tools including ALM risk modeling, derivatives and hedges, and off balance sheet servicing to manage risk;
  • Building capital to navigate deregulation and open competition, the FSLIC crisis and ultimate industry shutdown, two FDIC insolvencies, numerous recessions, the Great Recession 2008-2009, and the recent COVID economic shutdown and recovery,

Credit unions have demonstrated the ability to establish appropriate capital levels to meet every risk itemized in the RBC rule plus dozens of other events and changes that no model could incorporate.

Board and management have managed their individual capital levels to correspond to their business plans and the local economy in which they compete. Now NCUA wants to override that proven practice with a rule that treats each asset’s risk the same wherever and however the credit union operates.

Tomorrow’s blog will document the proposal’s unprecedented financial and regulatory burdens.

Credit Union History for Understanding Today’s Cooperative System

America’s Credit Union Museum is collecting oral histories of system participants to help future generations understand their cooperative roots.

Fifteen videos are now posted. Interviews are from retired leaders such as Carroll Beach, Dick Ensweiller, Brad Murphy and John Tippetts. Persons still active include league presidents Tom Kane and Caroline Willard, and Sarah Canepa Bang, the senior policy advisor to NCUA Vice Chair Kyle Hauptman.

Episode 15: The Deregulation Era

My first contribution discusses deregulation. It describes how Ed, Bucky and I learned with credit unions in Illinois to navigate the disruptive economic changes occurring in the late 1970’s and early 80’s. We went to NCUA using these lessons on a national scale.

The talk is 24 minutes. If your time is limited, here are some topics to scroll to:

3:00 Where the cooperative model fits on America’s economy

5:00 Learning about regulation and credit unions at Illinois’s DFI

12:10 We take our experiences to NCUA

14:00 Communicating what deregulation was; why it worked

16:15 Upgrading the NCUA’s internal capabilities

20:20 the PennSq bank failure

The Value of Oral History

The museum’s initiative to record individual’s credit union experiences will be invaluable to visitors and scholars. They are easy and fun to listen to, especially if you know the characters.

Hearing these examples will stimulate interest in cooperative history; more importantly it can give perspective on today’s topics.

Deregulation was not a political ideology, strategic blueprint or onetime response to a changing economy.

In credit unions it was nothing less than building a better system of “cooperative credit in the United States.” It turned upside down the practice of government making everyday business decisions for credit unions.

Rather that responsibility was now in the hands of those closest to the members-management and boards.

Most importantly these changes were developed mutually with full dialogue and participation by all segments of the movement.

When Leaders Lack Confidence in their Organization

What would you think if you learned that Warren Buffet was shorting Berkshire stock? Or Elon Musk prefers driving a Lexus?  Or Jeff Bezos doesn’t want to test fly his Blue Origin Space capsule?

None of these situations is true.  And because the opposite is the case, observers’ trust in these leaders and their organizations is sustained.

A Credit Union Example

Seven years ago, in October 2015, NCUA over the objection of board member Mark McWatters, approved a final 424-page RBC rule. This was NCUA’s second attempt to impose this new reg which was as equally unsupportable as the first.  Both attempts were universally opposed by credit unions.

One of the rationales for the rule stated in the 2014 NCUA Annual Report was “the issuance in 2013 of new risk-based capital rules by the FDIC, the office of the Comptroller of the Currency and the Board of Governors of the Federal Reserve System.” (page 12)

Certainly, an impressive endorsement by banking regulators.  However, in September 2019 the FDIC with the full concurrence of the Comptroller and Federal Reserve removed RBC requirements for all community banks under $10 billion.  Did NCUA follow its peer’s decision? No, It plodded on, kicking the can down the road even though one of their primary justifications was gone.

What the Rule Says About NCUA’s Self Confidence

But there is another insight, besides bureaucratic obstinacy, to take from the final proposal.

The agency published a two-page summary — Risk Weights At a Glance –as the final summary of absolute and relative risk of every possible balance sheet asset. Three judgments are illuminating.

Credit unions investing in the capital of the CLF have 0 risk.  Since the CLF has not made a loan for over a decade, it suggests how the agency is thinking about the CLF’s role assisting credit unions in the future.

The FHLB’s do make loans to credit unions. To qualify for these, a credit union must buy stock in the bank. NCUA determined these stock purchases should be assigned a 20% risk weighting.

Even though no FHLB organization has ever failed, the agency believes there is still a small risk.  But it is nowhere near the risk of a credit union investing in a CUSO, which requires a 100-150% weighting.

But the most ominous risk is for credit unions’ 1% capital deposit in the NCUSIF.  According to the chart, the 1% deposit cannot even be counted as an asset.  It must be subtracted in full from the numerator of the credit union’s net worth and from the denominator’s total of all risk weighted assets.

It is counted as having no value despite having been untouched for almost 40 years.  It is an earning asset, withdrawable in a voluntary liquidation or conversion to private insurance. On both credit union and NCUSIF balance sheets it is carried at full value.  Multiple national accounting firms have stated this asset “fairly presents” both aspects of this transaction.

What would subtracting this asset mean for the NCUSIF’s Risk Based Capital ratio!  If credit unions cannot count this as an asset, how can NCUA include these deposits in the NCUSIF’s net worth?

One interpretation is that this is just one of many foolish aspects of the final RBC rule which becomes effective January 1, 2022. But there may be more intention than one might think.

A Scary Thought

This NCUSIF total write-off of the 1%  from net worth, like the hypothetical made up examples first above , points to an uncomfortable reality.  This is an agency whose leaders lack confidence when managing the ever growing resources credit unions provide.  And if they lack the understanding of this cooperative fund’s operations, what message is sent to credit union members?

Today the NCUSIF equity level above the 1% deposit totals over $4.7 billion.  Should a loss of that magnitude or more occur, the primary question will not be about the status of the 1% deposit, but where was the regulator?

The cumulative loss rate for he NCUSIF over the past 12 years and two financial crises, is 1.5 basis points.  To project a loss at least 20 times this recent real world experience, is deeply troubling. (2,000 percent, i.e. 30/1.5)

That potential accountability is why the agency wants to eliminate the 1% from credit unions’ net worth today.  NCUA wants to avoid explaining how its oversight allowed such a situation to develop.

Now that is a scary thought.




Your Credit Union’s ROA and NCUA’s NOL Request

What if an NCUA proposal could reduce your credit union’s ROA from 3 to 8 basis points per year in perpetuity.  Would you care?  Would you even respond to NCUA’s proposal that had that implication?

This could be an outcome should credit unions fail to reply to NCUA’s request for comments on the policy process for changing the NCUSIF’s Normal Operating Level (NOL).  The continuation of NCUA’s NOL financial modeling incantations and Chairman Harper’s desire to remove NCUSIF premium limits have one goal: collecting more credit union funds.

Comments due by July 26 can be filed here. Or mailed to: Melane Conyers-Ausbrooks, Secretary of the Board, National Credit Union Administration, 1775 Duke Street, Alexandria, Virginia 22314-3428

Three Urgent Points in Response

In previous articles I have documented how multiple NCUA decisions managing the fund have shortchanged credit unions.  To return NCUA’s oversight of the fund to the best version of itself, three critical changes are needed.

  1. Convert all NCUSIF accounting to private GAAP from federal accounting principles. This should require the recognition of the full 1% deposit credit union contribution payable simultaneously with the reporting of insured shares when calculating the NOL. Also, all assets under management in AME’s should be audited and their financial data included.
  2. Discontinue NCUSIF financial decisions based on artificial models using hypothetical assumptions entirely disconnected from actual economic data and NCUSIF performance, which are then “projected” years into the future.
  3. Develop better management tools to forecast the current yearend NCUSIF outlook based on actual numbers and estimates from real events. The goal should be to manage the NOL in the normal range of 1.2 to 1.3.

Dividends should be paid when the fund exceeds 1.3.  If unusual events such as a period of historically low interest rates necessitate a premium, that should be temporary to remain within the normal NOL range.

Credit unions should oppose NCUA’s current financial conjuring process for the NCUSIF’s NOL. Manage to the normal range. There are no objective facts supporting a change to the longstanding legal guardrails on the NCUSIF structure.

The Origins of the NOL Change

The questions in the Board’s May request show NCUA’s intent to continue modifying the Normal Operating Level cap in place since 1984. This first change was made in 2017.  That board decision was based on financial fairy tales.  The purported models used neither actual economic events nor 33 years of audited NCUSIF performance outcomes to change the longstanding 1.3 upper limit.

This 2017 decision occurred in the eighth consecutive year of positive economic growth and credit union stable performance. It was simply an opportunistic money grab to avoid the political embarrassment of a premium.

NCUA fabricated numbers to approve a 1.39 NOL four years ago so that the NCUSIF could keep more of the $2.6 bn surplus when merging the TCCUSF.  If not raised above 1.3, additional dividends would have to be paid as required by the FCU Act.

Both board members publicly stated their approval was intended to retain these extra funds to avoid assessing a premium for the planned liquidation of the taxi medallion conserved credit unions.  Most credit unions objected to the proposal, but NCUA did not modify a single number.

Another proof of this financial artifice is that prior NCUA boards saw no reason to modify this normal limit during or immediately after the 2008-10 Great Recession.  The reason is simple, the NCUSIF worked as designed.

Carrying Forward a Fabrication

NCUA suggests continuing this financial fabrication process indefinitely.

The request for comments specifically lists criteria to “invent”  moderate/severe recession scenarios with characteristics based on NCUA’s judgment, using “qualitative” factors, and extending these assumptions out five years, more or less. The most important factor of the fund’s financial “drivers” is omitted –the agency’s operating expenses from the Overhead Transfer Rate.  In the past 13 years these expenses exceeded insured losses by $272 million.

This parody of a  financial model was used at the December 2020 board meeting to justify retaining a 1.38 NOL. That staff presentation projected a decline (loss) in the NOL of 16 basis points in a “moderate” recession.

This “forecast” was given in the same year as the worst one quarter fall of GDP ever recorded due to the COVID economic shutdown in 2020.  Despite this severe economic shock, the real NCUSIF outcome was that net cash recoveries exceeded insured losses by $10 million this past year. The “model” shown in December forecasted that the opposite should have happened—a big loss not a gain.

During the Great Recession years of 2008-2010, the NCUSIF’s cumulative loss rate on insured shares was only 2.8 basis points, a fraction of the 16 bps loss projection shown in December’s board presentation.

Every question listed in the NCUA’s request has an implied assumption that a larger and more frequent adjustment of the NOL outside the traditional 1.2-1.3 range from 1984-2017 is necessary. The 36 years of actual NCUSIF performance shows that managing within this NOL range is more than adequate for all economic cycles.

Financial Reporting Erodes Trust and Transparency

What makes the request even more problematic is NCUA’s use of misleading data to obscure the NCUSIF’s financial condition.  NCUA states that NOL at 2020 yearend was 1.26%.  The actual audited reserves to insured shares ratio was 1.318%.  NCUA’s number fails to include the true-up of the 1% deposit on yearend insured shares–an accounting practice followed until 2001.

NCUA’s conversion to Federal GAAP standards in 2010 further confused the fund’s financial reporting.  The NCUSIF’s “federal” accounting presentation of operating results is today converted by staff into a private GAAP income statement format for public reporting. The “federal” balance sheet includes unrealized investment gains and losses in the Cumulative Results of Operations (aka reserves) and omits all “fiduciary” assets in the AME’s including the billions in the corporate estates.

Federal accounting standards are inappropriate for a cooperative fund owned by credit unions.

The NCUSIF must return to private GAAP accounting principles for transparency, the same practice required for all credit unions.  It is the standard used for NCUA’s other three funds: the Operating Fund, the CLF and Community Development Revolving Loan fund

The 36 years of actual NCUSIF performance shows this cooperative design works in all economic cycles and industry conditions.

Not a Resource Challenge

This ongoing effort to permanently change the fund’s long proven NOL structure inverts the real challenge facing the NCUSIF model; the shortfall is not financial resources, but how the fund’s problems and capital are managed.

The largest natural person losses are not the result of a macro economic cycle.  The losses from the $239 million St Paul Croatian and the $40 million member shortfall in CBS Employees FCU were frauds carried out over many years, or decades.

The NCUSIF $750 million expense to resolve the taxi medallion conservatorships is an example of a fire sale to be rid of problems versus creating more cost-effective resolution options.  Why expense $750 million to end its responsibility for taxi medallion borrowers when its own conservators reported shortfalls of only $250 million in the credit union’s final call reports?

Write Now to Sustain NCUSIF’s Financial Soundness

For four years, NCUA has kept this NOL pantomine alive.  Meantime, the arc of NCUSIF performance continues creating a significant cooperative advantage for the system versus the FDIC’s model.

So far the only posted comment is from Mid Oregon CEO Kevin Cole.  He states the key issue clearly in his close:

For many credit unions, participation in the NCUSIF is the biggest, largely unmanageable risk they face. . .Because this risk is tied to actions of the NCUA Board and ultimately, other credit unions, individual credit unions have limited tools to manage the risk. . 

 The NCUSIF through the actions of the NCUA Board has the authority to assess credit unions’ premiums as needed to restore the fund’s equity ratio as needed. For this reason, it is not necessary or desirable for the NOL to be any higher than it absolutely needs to. Share insurance fund reports indicate that most credit unions are well capitalized and pose little risk to the NCUSIF. The credit unions with higher risk levels to the NCUSIF appear to be properly identified and working towards resolution, as evidenced by the low number of failures that cost the fund.

 For the simple reason that the NCUSIF has the authority and the ability to assess credit unions as needed, and credit unions have the means to pay, it makes no sense for the NCUSIF to hold more equity than legally required, except for identified probable losses. The capital belongs to credit union members and should be allowed to be deployed as those members wish.

Please add your voice to Kevin’s.






The NCUSIF Tool Every Credit Union and NCUA Should Use

The cooperatively created NCUSIF is an insurance resource designed for all economic seasons.  Whether there are clear financial skies, clouds or even once-in-a-hundred years storm (e.g. the COVID economic shutdown in the 2nd quarter 2020) the structure is flexible, transparent and proven.

Most importantly, its current performance is easy to monitor, by anyone.  It does not require a PhD in economics, 20 years of regulatory experience or even the expertise of a Black Rock.

There are only four moving parts.  There are 36 years of actual data to call upon.  Each component has historical and contemporary data to employ in the model.

The model’s four variables are:

  1. The operating expenses taken off the top from NCUSIF income through NCUA’s OTR process;
  2. Insured losses–there is data for all economic cycles. In the past 13 years and two crises, the cumulative rate is loss rate is 1.51 basis point for each $1 of insured savings.
  3. The yield on NCUSIF’s portfolio reflects the staff’s investment decisions, and is semi-fixed in the short term due to its current weighted average life of over three years.
  4. Credit union insured share growth. It is reported quarterly.  The most recent 13 year CAGR is 7% even including 2020’s unusual rate of 22%.

The Model Demonstrates A Better Way

In 1984 credit unions and NCUA worked together to create a cooperative fund that would be A Better Way than the failing premium-based systems. This dynamic model shows how this financial design works and more importantly how it can be even better managed today.

The model has two key capabilities.  It dynamically links the four financial variables so that any input change automatically recomputes all outcomes.

Second it translates the variable input measures into understandable, actionable data.  Converting historical insured loss experience (bps), rates of share growth and investment yields(%), and operating expense ($) into one integrated financial design ensures the model always aligns with the size of insured risk.  For example 1 basis point loss in 2010 is $76 million; however in 2020, that same loss rate is $144.5 million due to the growth of insured shares. The model’s output is provided in all three measures: $, % yields, and basis points to easily understand the options for the yearend NOL.

Events have demonstrated that NCUA’s NCUSIF models used in 2017 through today are not based on any objective reality.  The so called scenarios are financial myths developed to support NCUA’s desire to remove the guardrails on the most successful deposit fund ever managed by the federal government.

The Dynamic Spread Sheet

This is the link to view the xcel calculator with the numbers used below.  To enter your own numbers, feel free to make a copy (open this view-only copy with google sheets, click File, then Make a Copy).  Using your copy will automatically calculate the yearend outcome with your inputs.

Models are not answer machines. They inform judgments.  They require objective validation.  When used properly, they identify options and improve management effectiveness. The following is the example shown in the “view-only” model link above.

An Example With Actual Data

The audited yearend NCUSIF data from December 31, 2020 is entered in the view-only example. This includes the audited reserves ($4.7 Bn), insured shares ($1.467 Tr), the reserve ratio (.318) and a 2021 yearend goal for this ratio (.30).  With the 1% credit union deposit, the spread sheet calculates the performance required to reach a 1.3 NOL at 2021 yearend.

The columns E, F, G and H are four variables that can be updated anytime.  I have entered both current numbers and informed estimates.  For share growth I used the first quarter’s annual rate of 23%; for loan loss I entered .5 of a basis point which is below the long term 1.51 rate because to date, the NCUSIF has reported net recoveries.

For total investments I have added the increase in 1% deposits from the year end true up and a mid-year estimate.  However, the portfolio’s size is not increased 100%, reflecting that these additional deposits will only earn for 9 and 3 months, respectively.  The $190 million expense total is higher than 2020’s expense and uses the 2021 budget modified by actual results through May.  Finally, for the yield, I entered NCUSIF’s current year-to-date portfolio return, 1.27%.

The five green columns show the projected yearend outcome from these inputs.  The required portfolio yield is 5.28%, way above the current yield.  This shows a yearend NOL of 1.257.  To raise that outcome to the target 1.30 level would require a maximum premium of 4.33 basis points of 2021 yearend insured shares or $782 million—if nothing changes in these input assumptions.

That certainly seems a dour result in this relatively stable and growing economic climate.  It does not necessarily require a premium. At numerous points in the past when the NOL ended below the 1.3% cap, NCUA has foregone a premium at yearend, most recently in 2016 ending  with a 1.24 NOL.

A Different, More Probable 2021 Scenario

However, this forecast and assumptions are an unlikely outcome. Extraordinary share growth is driving this result.  The rate is already slowing.  June and September call reports will provide a more relevant number.

For example, if insured shares grow at the 13-year average of 7%, then the results are totally different.  The required portfolio yield would be just 1.6%, and the year-end NOL, without any premium, is 1.296, almost at the 1.30 cap.   Further, if there were no insurance losses (zero loss provision), and only continued recoveries, the yearend ratio would be 1.308, above the NOL target.

It’s Not Rocket Science

Credit union CEO’s and boards are living daily the operational experiences underlying these inputs.  They know industry growth trends, delinquencies and certainly have a feel for interest rates.  Their input assumptions can be informed by their real world understanding, not guesses about the future.

This simple model should empower every credit union to evaluate the information and NCUSIF projections used by NCUA. Responses should be formed from their expertise and data, not unsupported opinion.

But more importantly, anyone who uses it will understand the flexibility and viability of the NCUSIF’s cooperative design.

Every year the NCUSIF automatically grows at 80% of the share growth rate via the 1% deposit true-up (1/1.25).

If insurance losses, share growth or yields fall outside the long-term credit union system’s experience, then a premium is an option to maintain a targeted reserve ranging from .2 to .3.  There were many adjustments to the NCUSIF equity during the 2008-2020 period.  However, the two premiums in 2009 and 2010 averaged only 1.3 bps annually over these 13 years.

The Yield Calculator as a Management Tool

Another value of the NCUSIF model is using the Yield Calculator and the recent 13-year averages to calculate the breakeven yield on investments needed to maintain an NOL target, say 1.25, or within a narrow range.  One must first convert the operating variables to basis points. Adding the 7% insured share growth rate(1.75 basis points) plus 1.51 basis point loss experience, and the 1.6 bps of operating expenses totals 4.86 basis points.  Dividing this total by the portfolio size of 1.25  (the midpoint of the  NOL)  shows the “breakeven” investment yield needed to support this NOL is 3.9%.

The model clearly highlights the relative importance of each of the four variables.

Operating expenses come off the top of revenue. If the OTR had remained aligned with the proportion of state charters, or 50%, the operating expense share would have been 1.41 (not 1.6) basis points, thus lowering the required breakeven yield.

If share growth were 22%, not the long-term average of 7%, the bps for this variable would increase to 5.5 and the required yield jumps to 6.9%.

 Managing the NCUSIF’s Portfolio

The model demonstrates the importance of monitoring the decisions about NCUSIF’s investment  portfolio. If the current yield 1.27% and 4.86 basis point of inputs were frozen forever, credit unions might have to pay an annual premium of 3 basis points (1.27%-3.9% equals yield shortfall of 2.63%/.8/1.07 = bps premium on the new year end 1% insured share base) to maintain the NOL at a 1.25 breakeven level.

Informed oversight of the NCUSIF’s portfolio decisions is a vital responsibility of NCUA  management and board.  Market rates change. In the interest rate trough of this economic cycle, should NCUA be making 8-year fixed term investments for a 1.26% yield (April 2021) when knowing the breakeven goal for NCUSIF is  3% to 4%?

On November 18, 2018, the yield on the two-year treasury bill was 2.98%.  Today the yield is .26%.  If NCUA’s investment managers are supporting the NCUSIF’s financial model, should they be routinely  filling out an investment ladder up to 8 years in this part of the economic cycle? Common sense says no—this will reduce NCUSIF’s income, shortchanging credit unions years into the future.

The economy is at an historical low point in this interest rate cycle.  The dominant economic topic today is the inflation outlook.  Are price increases here to stay or a transitory event?  Whatever the outcome, realistic judgment suggests that rates will be higher 6-12 months from now–the only question is how much higher. An example of this change prospect  from the past ten years is that the peak in the 5-year treasury yield was 3.07% on October 1, 2018, just two and a half years ago.

Tools Are Only as Good as the User’s Skills

Every credit union uses spread sheets.  This model is a simple, automatic xcel tool-just add your own judgment.  All the variables are in the formulas.  Some inputs are set; others easy to project.

Interpreting the outcome(s) is the art.   Bias will sometimes cloud these judgments. That is why it is vital that commentators on NCUA’s request for NOL comments document opinions with factual underpinnings and an understanding of how special this coop fund is.

Good luck with the model; critiques/questions are welcome.  Tomorrow I will suggest points to make by credit unions responding to this NCUA’s NOL request.

Facts vs. Fictions: the NCUSIF Financial Model Works—Even When Mismanaged

This is the third of five articles looking at how the NCUSIF financial model has performed during the past 13 years, even the times when it has been mismanaged.

NCUA’s request for comments on NCUSIF’s NOL assumes the core model of a .2 to .3% level of retained earnings on top of the 1% credit union underwriting is inadequate. Chairman Harper openly states this view. The facts going back to 1984, or 36 years of operation, show otherwise.

From 1984 through 2007, the NCUSIF’s financial model delivered the results promised. Losses were minimal, there were dividends in a number of years; premiums assessed only twice; and the NOL easily maintained in the .2 to .3 range.

So successful was the model that the fund paid six consecutive dividends from 1995 to 2000 and again in 2006 and 2008. It would have distributed more except NCUA changed the way it calculated the NOL by excluding the required yearend true-up of the 1% deposit beginning in 2001.

The following graphs are based on NCUSIF’s audited statements for the years 2008 though 2020. This 13-year span is used because it includes the two great post-Depression systemic financial crises and years of historically low interest rates to prod recovery. Here is what the data shows.

Coverage for Insured Loss Risk

Throughout this period, the cumulative actual losses were 1.51 basis points of insured shares. In five of these years, losses were less than 1 basis point. In 2020 cash recoveries exceeded losses.

During the 2008-2010 Great Recession actual losses never exceeded 3.5 basis points. The peak year in losses was 2018 at 7 basis points, when NCUA liquidated the taxi medallion conserved credit unions.

The year-by-year actual loss and the cumulative long term loss experience of 1.51 basis points are shown below. The normal NOL range of 10 basis points is almost 7 times greater than this historical average.

However, NCUA’s accounting estimates for loss provision expense, on a year-in year-out basis, have no correlation to actual events. This provision expense is shown in annual reports as “insured losses” and creates a very misleading presentation.

The graph below shows that the provision account has been at times funded at a level 10 times (1000%) greater than the following year’s actual losses. In credit unions, the allowance account ranges between 1 and 1.5 times actual losses.

The result of this misleading loss provisioning results in dramatic swings, from an expense of $737 million in 2010, to a reversal of $526 million in 2011, a total of income statement impact of $1.3 billion in just two years.

The graph below shows how these mistaken estimates create wide variance in reported net income versus the actual insured losses. These exaggerated loss provisions were the basis for assessing unnecessary premiums rather than informed judgments based on historical loss experience and a common sense view of current events.

NCUSIF Operating Expenses Exceed Insurance losses

Over the past 13 years, NCUA has charged the NCUSIF $2.2 billion in expenses, an amount $270 million greater than all insured losses of $1.9 billion for the same period.

The result is that NCUSIF has become the primary funder of NCUA’s operations, rather than the operating fee intended for this purpose.

The graph below shows that in eight of the past 13 years, operating expenses have been greater than insured losses:

This outcome is because over 93% of NCUSIF’s operating expenses are from the Overhead Transfer Rate (OTR). This transfer charge has ranged from a low of 52% (2008) to as high as 73.1% (2016) of the agency’s total yearly spending.

The combination of these arbitrary changes in the OTR (State chartered federally insured credit unions have never exceeded 50% of the insured base) and wildly inaccurate loss provisions, illustrate the difficulty of relying on NCUA’s financial reports to assess the sufficiency of the NCUSIF model.

In fact, a reasonable interpretation is that the model has functioned well despite significant missteps by NCUA.

NCUSIF’s Total Yearend Reserves are Much Higher than the Reported NOL

Since 2001, NCUA has not included the yearend true-up of the 1% required credit union deposits when calculating the NOL. The result is to understate the actual NOL. It is easy to calculate the actual reserves above the 1% deposit contributions by dividing the reserves by insured shares.

When this is done as shown on the chart below, the actual NOL is always greater than the number NCUA reported. In some years the difference is over 5 basis points.

For example, in 2020 NCUA reported the NOL was 1.26% whereas the actual number was 1.32%. Each basis point of 2020 insured shares is $144 million. This belated true-up practice substantially misrepresents the NCUSIF’s actual financial condition.

Moreover, by adding the allowance for loss (already expensed) to the retained earnings, the actual dollars and margin in bps for covering losses is greater than the just ending NOL. As shown below, the reserves plus allowance account have exceeded .3% level (or legal cap for a premium) for the NCUSIF every year since 2008. In 2017 this total reached a peak of 55 bps or 25 bps higher than 1.3%.

Federal GAAP Accounting Further Confuses NCUSIF’s Presentation

The NCUA board’s adoption of Federal GAAP standards for NCUSIF reporting in 2010 created two additional transparency problems.

Federal GAAP does not report retained earnings/reserve in the same manner as private GAAP. The reserves called Cumulative Results of Operations include any change in unrealized losses or gains on the fund’s investment portfolio, its largest asset.

The graph below shows how this “valuation” over-or-understates actual reserves by as much as $466 million (2020). NCUA does not use this valuation when it calculates the NOL, but it is the amount shown in NCUSIF’s “net position” in the balance sheet prepared using Federal GAAP.

A second point of confusion is adjustments to prior periods of loss estimates for NCUA’s liquidation estates. As fiduciary assets, these amounts are not included on the “government entity’s” (NCUA’s) balance sheet. Only the net change in receivables is shown.

The chart below shows these prior adjustments to loss estimates total $313 million and are increasing in amounts. These managed assets should be audited, as in private GAAP and not excluded, so these valuations are subject to independent review.

The Resilience of the NCUSIF Model

The data show the NCUSIF financial model is resilient. The combined total of insured losses (1.51) and operating expenses (1.77) are well within the 10 basis point traditional NOL range. This is true even in this 13-year period which includes two major economic crises. This is the actual outcome, not estimates, forecasts or misleading modeling outputs.

To pay this combined expense of 3.28 basis points requires a 2.5 % investment yield (3.28/130) on the portfolio. If insured losses are 0, then the yield for operating expenses alone is only 1.36%. A portfolio yield range of 1.36% to 2.5%, on average, is sufficient to provide revenue for these two costs.

Incorporating Share Growth in a Dynamic Model

However additional income is necessary to maintain the NOL level in line with insured share growth. Tomorrow I will attach a dynamic spread sheet that shows how to easily calculate this required yield incorporating all four variables. Anyone can update and project their forecast for yearend NOL. As a preview, the 13-year insured share growth is 6.9%. Therefore the .30 in reserves would have to grow by this rate to maintain the ratio’s current level, which would require 1.6% additional yield. (2.1/130).

If the retained earnings fall below the traditional 1.3% NOL, then the board can assess an annual premium to this cap if needed. Historically, the board has rarely used the premium just to raise the fund’s balance to the maximum 1.3% cap. Rather, it has left the NOL in a range of 1.25 to 1.29 or wherever the financial results ended.

NCUA’s Calculations Raising the NOL are Suspect

When merging the TCCUSF into the NCSIF in 2017 and again in the December 2020 Board meeting, staff recommended keeping the NOL above the historically proven 1.3% cap. The board did not have this authority until CUMAA passed in 1998 which raised the maximum cap to 1.5%.

During the 2008/09Great Recession and afterwards, the board retained the 1.3% cap. The most accurate statement in the 2017 staff presentation for raising this cap was in the concluding slide: “Actual results may vary from projections.” These same models were again used in the December 2020 board NOL update even though wildly inaccurate compared with actual experience.

The 2017 NCUSIF projections were not due to any fundamental change in the NCUSIF’s financial capabilities or the economic outlook. Rather it was an opportunistic effort to retain as much as possible of the TCCUSF surplus upon merger. These funds above 1.3% would then facilitate the liquidations of two taxi medallions in conservatorship without assessing a premium. Board member Rick Metsger and Chairman Mark McWatters confirmed this in public comments.

NCUA staff continues to use this strained, artificial logic from the TCCUSF merger to keep the NOL above 1.3. Lifting the NOL cap in 2017 was opposed by credit unions unanimously in their comments. But NCUA changed not a single number in their plan. Credit unions were also clear that NCUA should go back to the 1.3% cap as soon as possible if the Board approved 1.39 as the new NOL cap.

In December 2020, staff recommended a 1.38 NOL cap with the same hoary reasoning. They projected a 16 basis point decline of equity in a moderate recession and another 2 basis points from lower corporate recoveries, adding 18 basis points to the 1.2% lower NOL limit.

As in 2017, these modeled numbers are not based on any verifiable data. Why a lower than expected recovery on corporate AME claims reduces NOL is not explained. Projecting a decline in NOL to justify raising the NOL is circular logic.

The staff’s model includes no recognition that the largest expense to the NCUSIF is NCUA’s OTR operating costs. Staff did not use this expense factor when justifying the 16 basis points needed to “prefund” NCUSF equity for a moderate recession.

The NCUSIF is already funded by the ongoing 1% credit union contributed capital.

Dystopian Future Forecasts

NCUA’s models are fictions. Assumptions are not back tested against real world results.

Dressed in real numbers, projections are presented as “future facts” (scenarios), which are unknowable. However these financial myths have immediate consequences for credit unions. The Board bases present day decisions such as OTR expense transfers, premium assessments, capital spending or dividends on these hypothetical multi-year forecasts. Credit unions then pay the price now for incorrect projections which will only be revealed in the future.

NCUA’s dystopian forecasting was responsible for the corporate catastrophe. It led to an incorrect understanding of reserves, underestimated future earnings from investments, and ignored the improving current trends in market valuations.

For natural person credit unions, the most dramatic example of mistaken forecasts is NCUA’s $1.4 billion loss provisions in 2009 and 2010. Actual cash losses for those two years were $373 million. The difference of over $1 billion came out of credit union pockets via premiums.

The year of the worst GDP quarterly fall in American economic history was 2020. Yet, the NCUSIF reported positive cash recoveries, not insured losses. This short lived recession did not impact on equity as projected in 2017. Yet staff still asserted the need for a 16 basis points margin even as the 13-year cumulative insured loss experience declined further to 1.5 basis points.

The dominant issue for credit unions when commenting on NCUA’s NOL request is straight forward. The data show the challenge is not adding additional resources, but rather managing more effectively and transparently those already in place or on call

Tomorrow I will link to a simple spread sheet anyone can use to project outcomes for the NCUSIF. The assumptions entered and data are clear. Forecasts can be updated with real data at any time. The prospect for a premium or dividend is easily seen. A breakeven yield to achieve an NOL target is instantly calculated.

NCUA’s NCUSIF Accounting Short-Changes Credit Unions

This is second of five articles is to assist credit unions responding to NCUA on the capital adequacy (NOL) of the NCUSIF due July 26.  The first article quoted Chairman Todd Harper’s unsubstantiated view that  NCUSIF’s structure is inadequate and requires more NCUA authority to assess premiums.

This article reviews the accounting changes, beginning in 2001, that reduced NCUSIF dividends and increased expenses. These changes have prompted some to suggest that NCUSIF’s financial design is inadequate — a mistaken judgment I will challenge.

NCUA published NCUSIF’s audited financial statements for 2008 and 2009 only after a prolonged delay. With these audits NCUA changed accounting standards creating  confusion, misleading presentations and uncertainty about what assets were audited.

The NCUSIF is unique due to its cooperatively underwritten financial structure.  To provide relevant responses to NCUA’s request requires agreement on basic facts.  NCUA’s changes  in 2000  deviated from the NCUSIF’s prior consistent accounting practices used since the 1984 redesign was approved by Congress.

These changes resulted an ever increasing draw on the NCUSIF to pay a larger proportion of NCUA’s operating expenses and underpaying dividends to the credit union owners.

Since 2008  and two financial crises, the data show that  the NCUSIF’s  operating expenses exceed  insurance losses from problem credit unions. Instead of a capital reserve helping credit unions, the NCUSIF has become the main source for financing the agency’s administration, not the required operating fee.

The following is a description of these significant changes in NCUA’s management of the NCUSIF.

Manipulation of the NOL

From 1984 though 2000, NCUA was consistent in its calculations of the NOL. Credit unions’ 1% deposit funding obligation has always been an explicit legal liability. As stated in the Act: Federally insured credit unions are required to maintain a deposit equal to one percent of their insured shares with the Insurance Fund. 12 U.S.C. 1782(c)(1)(A)(i).

In the NCUSIF’s 2000 Annual Audit, this 1% capitalization and NOL calculation are reported as follows in audit footnote 5.

“The Credit Union Membership Access Act (CUMAA) of 1998 mandated changes to the Fund’s capitalization provisions effective January 1, 2000. . . The NCUA board has determined the normal operating level to be 1.33% as of December 31, 2000 which considers an estimated $31.9 million in deposit adjustments to be billed to insured credit unions in 2001 based upon insured shares as of December 31, 2000. . . The CUMMA mandates that the use of year-end reports of insured shares in the calculation of the specified ratios, and the dividends related to 2000 will be declared and paid in 2001 based on insured shares as of December 31, 2000.”

This 1.33% ratio was calculated by dividing NCUSIF’s audited reserves plus 1% of yearend insured shares by total insured shares. This method was the basis for sending credit unions a sixth consecutive dividend from 1995 through 2000.

But in 2001, the Board changed this calculation both retroactively for 2000 and going forward in 2001 and ever after. This change is described in footnote 5 in the 2001 audit:

“The NCUA Board has determined that the normal operation level is 1.30% as of December 31, 2001 and December 31, 2000. The calculated equity ratio at December 31, 2001 was 1.25%. The equity ratio at December 31, 2000 was 1.33% which considered an estimated $31.9 million in deposit adjustments billed to insured credit unions in 2001 based upon total insured shares as of December 31, 2000. Subsequently such deposit adjustments were excluded and the calculated equity ratio at December 31, 2000 was revised to 1.30%.

However, NCUA did pay the previously calculated dividend for 2000 based on the initial 1.33% NOL. “Dividends of $99.5 million which were associated with insured shares of December 31, 2000 were declared and paid in 2001.”

If the same method of calculating equity to insured shares ratio in 2000 were applied in 2001, the resulting NOL would be 1.303%– not 1.25%– thus triggering a small dividend.

By undercounting the full 1% deposit NCUA avoided paying  a dividend. It misstated the actual NOL of 1.3% by 5 basis points, a significant reduction.

There was no basis for this change. In every year prior to 2001 NCUA sent invoices after yearend. That is the case today as NCUA bills credit unions their required true-up on insured savings after receiving the yearend call reports.   The  yearend NOL determination is easy and transparent– both the audited reserves and credit unions’ report of insured shares are available at the same time from the same source.

In making this arbitrary change to the NOL calculation, NCUA has understated the actual NOL to the present day. The underreporting of this ratio meant NCUA did not pay dividends as required (as in 2001) and understated the actual NOL ratio.

Graph Heading;  NCUA’s Reported NOL Understates NCUSIF’s Actual Capital Ratio (2008-2020)

Changing the OTR in 2000

Another NCUA draw upon the fund began at this time. In 2000 NCUA increased the percentage of its operating expenses charged to the fund via the Overhead Transfer Rate (OTR). The change from 50% to 66.7% was a 33% increase in just one year. This increase occurred  even though state chartered federally insured credit unions (FISCU’s) were only 44% of the total 10,316 of the NCUSIF’s insured base.

The agency continued to use this annual transfer in uneven and undocumented patterns reaching a peak OTR of 73.1% in 2016. This  increase reimbursement for agency expenses and understating the actual NOL resulted in no or only partial dividends due the credit union owners.  This dividend drop-off took place even though insured losses from 2001 until 2008 were either 0 or less than 1/2 of a basis point.

For example, in the 2006 audit, the NCUSIF declared an NOL of 1.304 requiring a dividend of $51.5 million. The actual ratio was 8 basis points higher and should have resulted in $103 million more paid out but which was kept in the NCUSIF.

No Timely Numbers and Changing Auditors

In the midst of the 2009 financial crisis, NCUA conserved US Central and WesCorp. Chairman Michael Fryzel asserted in a March 21, 2009 Wall Street Journal interview, “With us in control, we’d get honest numbers.”

That is exactly the opposite of how NCUA reported its own numbers for the NCUSIF. All the corporates routinely filed and published full 5310 monthly financial reports with current portfolio valuations within 30 days of every month end.  The corporates were managing over $100 billion in investments. But  NCUA did not release its December 2008 NCUSIF audit until a year and a half after yearend.

During this period of  uncertainty, NCUA took the following steps:

  • It replaced the NCUSIF’s 2008 auditor Deloitte, Touche with KPMG for 2009;
  • It changed auditing standards–from private GAAP to government GAAP;
  • It released the December 2008 audit on June 10, 2010, 15 months after the statutory April 1 deadline for reporting to Congress.

These actions typify a reporting entity with serious managerial difficulties subsequently noted by both auditors.

By delaying the release of the numbers, credit unions did not know the status of the fund. NCUA statements ranged from Chairman Fryzel’s assurance in the WS J that “regulators aren’t concerned about the health of any other wholesale credit unions besides the two brought into conservatorship” to wild exaggerations of losses that would cause write off of credit unions’ 1% deposit base.

The corporates managing the problem assets reported timely; NCUA did not.

NCUA provided few factual updates to counter the rampant hyperbole. Meanwhile the economy showed positive growth in GDP and market valuation recovery beginning in the fourth quarter of 2009.

Without NCUA’s numbers, credit unions were in the dark about how a mutual solution might be developed to minimize loss using the collaborative financial tools of the CLF and NCUSIF.

Late Reporting and Changing the NCUSIF’s Accounting Standard

Until 2009, “the NCUSIF historically prepared its financial statements in accordance with accounting principles generally accepted in the United States of America (“GAAP”), based on standards issued by the Financial Accounting Standards Board (“FASB”), the private sector standards setting body.”

The issue following this standard was whether US Central and or Wescorp, conserved in March,  must be consolidated with the NCUSIF under the Variable Interest Entity (VIE) rule. As a subsequent event and prior to the completion of the December 2008 audit, the VIE presentation needed to be resolved. As stated in their audit, this situation was one indication that NCUA did not fully grasp the consequences of their actions.

The resolution of this was described in the audit notes for 2008:

It was concluded that for 2008 the NCUSIF would be the primary beneficiary of certain identified VIEs based on variable interests held by the NCUSIF at December 31, 2008, and therefore, the NCUSIF would have been required to consolidate such VIEs in its financial statements for the year ended December 31, 2008. However, based on the actions discussed below, it was concluded that the TCCUSF would be the primary beneficiary of these same VIEs based on variable interests held by the TCCUSF at December 31, 2009.

The shift in primary beneficiary from 2008 to 2009 was the result of the June 18, 2009, actions of the NCUA Board to transfer the legal obligations related to CCUs from the NCUSIF to the TCCUSF. Such actions relieved the NCUSIF for the costs and related obligations of stabilizing the CCU system, as provided by Public Law 111-22, which was enacted May 20, 2009.

This change in reporting entity has been applied retrospectively to 2008. Accordingly, the accompanying financial statements for the year ended December 31, 2008 do not reflect the consolidation of any CCUs.

Avoiding Private Auditing Standard Requirements

To avoid VIE accounting requirements, NCUA in its September 16, 2010 public board meeting adopted federal GAAP.  These are excerpts of this discussion:

Mary Ann Woodson, the agency’s CFO at the time: The purpose of this action is to request Board approval for the National Credit Union Share Insurance Fund to adopt accounting standards promulgated by the Federal Accounting Standards Advisory Board, also known as FASAB. These standards are also commonly referred to as Federal GAAP.

The Share Insurance Fund currently applies Financial Accounting Standards Board, or FASB, standards. These accounting standards are used by commercial businesses in keeping their books and records and in preparing their financial statements.

On June 17, 2010, the NCUA Board adopted FASAB accounting standards for the Temporary Corporate Credit Union Stabilization Fund. Since then, we have gained more experience with FASAB and we have seen firsthand that FASAB standards more appropriately meet the financial reporting requirements of the NCUSIF and its stakeholders. Also, FASAB is the preferred standard for federal entities. . .

Debbie Matz, then NCUA Chair: So, if we had switched to this before last year, we wouldn’t have had the long drawn-out issue with the auditors that required us to meet with FASB in order to get a clean audit opinion at the end of last year, but it would have eliminated the inconsistencies and vagaries that created that situation?

Mary Ann Woodson: Certainly it would have helped. Yes.

Debbie Matz: Well, then I support it. Thank you.

Both auditing firms commented that NCUA did not understand or follow basic auditing procedures whichever standard was in place:

The 2008 and 2009 NCUSIF audits  released on June 10 and 11, 2010 included auditor’s comments on accounting deficiencies. The 2008 Deloitte audit reported “material weakness” in NCUA’s failure to “properly identify the appropriate accounting treatment under Financial Accounting Standards Board (FASB) Consolidation of Variable Interest Entities with respect to its variable interests in certain corporate credit unions.”

KPMG’s 2009 audit reported a “significant deficiency.”  The SIF, the report states, “does not have sufficient staff resources with the experience in technical accounting and reporting requirements that the entity requires to consistently perform certain internal control activities, particularly those related to the preparation and review of the financial statements. . . The SIF does not have procedures in place that require review and approval of the journal entries and related supporting documentation.”

Adopting a Misleading Accounting Standard

In this 2010 board meeting NCUA  adopted federal GAAP as the “preferred standard for federal entities.” But it did so only for the NCUSIF in order to avoid certain disclosures according to Matz’s question. However the Operating Fund and CLF continued and are still audited using FASB GAAP.

The problems adopting Federal GAAP are much greater than VIE disclosures. First the NCUSIF is not a federal entity, but rather a cooperatively funded capital reserve for the credit union system. There are no federal appropriations or funds involved.

Investments are in Treasury securities, but that does not make the NCUSIF a government entity any more than it would a credit union holding the same investments. The NCUSIF is totally privately funded and only for credit union system use.

The Federal GAAP income statements and balance sheets are confusing and misleading for credit unions. As explained by Woodson, the agency’s then CFO: “Under FASB, or commercial GAAP, the focus is on netting revenue against expenses for either a net income or a net loss. Under FASAB, or Federal GAAP, the focus is on the cost to government. So we start with the costs and reduce that amount by any revenue earned, to get to the total cost of operations, or said another way, the net cost to the government.”

Under FASAB the focus is on separating transactions within the government from those with the public.  But all NCUSIF transactions are with credit unions, that is, the public.

The FASAB balance sheet presentation and income are prepared to show these two activities, but are totally irrelevant to how the NCUSIF operates. The numbers presented in this manner are at best confusing and at worst, misleading.

For example, under Federal GAAP the fund’s retained earnings are described as the Cumulative Result of Operations. This description includes unrealized gains and losses on the NCUSIF investment portfolio which can  over or understate actual equity significantly.

In the NCUSIF fiscal 2020 results using Federal accounting, it reported a $500 million increase in its net position to a total of $5.1 billion. However, the actual net income from operations was only $32.9 million. This  number is nowhere in the Federal GAAP statements-but reported on Slide 3 in the NCUA’s staff December 31 presentation to the board using private GAAP. This substantial  difference in the increase in total equity of $400 million versus operating income of $33 million is due primarily to including additional  unrealized gains in NCUSIF investments from the prior year.

NCUA staff reports the total of the federal Cumulative Result in its slides.  But when calculating the NOL, staff eliminates any unrealized gains or losses  highlighting the confusion between the two standards when presenting financial performance.

The difference in NCUSIF’s actual retained earnings and “cumulative results of operations,” varies by hundreds of millions each year (2008-2020) as shown below.

However the most serious defect in federal GAAP is that  “fiduciary assets” are non-governmental and therefore not  on the NCUSIF’s balance sheet.  These are all of the AME’s  including the corporate estates which total  billions.  They are in effect off the books.

As stated at the top of the AME financials issued by NCUA, the amounts are unaudited.  Under private GAAP these amounts would be audited and included  on the NCUSIF’s balance sheet.  Currently the amounts for corporate and natural person AME’s are presented  in Federal GAAP as “net”  receivable assets subject to various accounting and income recognition rules.

Using the Proper Accounting Standard in a Consistent, Transparent Manner is Critical

To evaluate the NCUSIF’s financial design and capital adequacy, an essential first step is adopting private FASB accounting. This is how NCUA staff routinely presents the board with NCUSIF slide updates—except for the balance sheet entry showing “cumulative results of operations.”

Returning to this accounting standard will help all users more easily understand when monitoring the fund. It will:

  • Present financial performance in a standard balance sheet and income statement format readily understood and monitored by credit unions;
  • Restore the NOL calculation as was done from 1984 through 2000. The misleading recognition of the 1% deposit, citing a billing delay, is inconsistent with credit unions’ standing legal liability and misstates the fund’s actual NOL status at yearend.
  • Bring more rigorous consistency to estimates of loss provision expense. Since 2009 there has been no relation between the provision expense and actual cash losses.
  • Shed greater insight into NCUA’s decisions including the management of its investment portfolio, the AME’s and the probability of premiums or dividends at yearend.

These accounting distortions have contributed to questions about the fund’s sufficiency and flexibility in a low rate environment. This is an important issue in the current environment. But it can only be analyzed if there is a rigorous, objective and consistent presentation of performance.

The fund’s revenue is primarily dependent on the yield on the investment portfolio. But whether that will be a factor causing a premium depends on two other critical events:

  1. What is the expected level of actual losses to the fund (versus the seemingly arbitrary loss provision expense)?
  2. Will the NCUA’s ever growing  expenditures transferred to the NCUSIF be brought into better alignment with proportionate share of state insured credit union risk?

Tomorrow I will present the NCUSIF’s cumulative performance from 2008-2020 to address the question, does the unique NCUSIF design still work after the two worst financial crises since the Great Depression?