Covid-19 revenue and spending patterns fueled strong growth in public funds investment balances over the past several years, and recent Federal Reserve data show that investment balances continued to expand this year to $3.7 trillion as of March 31, 2023.

The data is buried in a 200 + page quarterly report on Financial Accounts of the United States from the Federal Reserve, and although it has some significant limitations—we’ll get to these in a bit—it is the best there is when it comes to accounting for public sector investment trends.

Here’s what we’ve seen in the past several years:

• Aggregate cash and investment balances grew by $855 billion, or 32%, from 2019-2021 as the Covid-19 lockdowns delayed state and local government spending, and Federal Covid relief funds swelled state and local government balances. Interestingly, the Pew Foundation Federal Pandemic Project estimated total Federal aid to state and local governments for Covid relief to be about $800 billion so this might account entirely for the growth.
• Balances stabilized in 2022 but continued to grow in the first quarter of this year. Investment assets grew about 3% in the first quarter of 2023. This is surprising. Though Federal Covid relief payments have largely ended, American Rescue Plan Act funds are being spent by states and localities, and overall economic growth is slow. Other factors—including high interest rates that generate much higher investment returns—swelled portfolios by $124 billion over year-end to $3.7 trillion. (The figure is not seasonally adjusted but 3% in a quarter is a hefty annual rate.)

The growth could be reversed in coming months as state and local governments continue to spend Federal Coronavirus relief funds before the 2026 deadline. And an economic downturn would depress tax and other revenue receipts and push up the costs of social and economic aid programs, drawing down investment balances.

Source:  Federal Reserve Release Z.1; PFII calculations

Portfolio quality improved. The growth in assets over the past several years provided portfolio managers with fresh cash to invest. Much of this went into Treasury securities. Total Treasury holdings represented 45% of assets at quarter-end. This is up dramatically from 31% in 2018, an increase of nearly $900 billion over the period.  Investments in Federal agency securities declined, perhaps reflecting diminished supply and smaller yield advantages.  Nevertheless, at quarter-end Treasuries and Agencies combined represented 55% of total financial assets, up from about 51% in 2018.

Source:  Federal Reserve Release Z.1; PFII calculations

Liquidity also improved. Bank cash and demand deposits plus money market fund assets added another 12% to holdings that can be characterized as highly liquid.

 Money market funds made up a small portion of portfolios in 2018 (less than 1%), and although their allocation nearly tripled, they represent only 2% of portfolios as of March 31.  (Local government investment pools are not accounted for in the Fed release.) 

Cash plus time and savings deposits represented about 20% of investment assets on March 31, 2023.   Amounts are higher compared with 2018, but the portfolio allocation—about 20% of investment assets—is remarkably stable. Interestingly the nature of bank deposits changed, with the portion in time and savings accounts in 2018 (75%) declining in favor of cash equivalents. This is further evidence that aggregate portfolio liquidity has improved.

Despite the overall increase in assets, total credit exposure (commercial paper, corporate bonds, and municipal obligations) was reported to have declined by about $37 billion, and in the aggregate represents just 7% of portfolios as calculated by the Fed. (Take these figures with a grain of salt, for the reasons we describe below.)

Bottom line:  Portfolio managers used the new money they received over the past five years to shift allocations to securities and cash with greater liquidity and reduced exposure to credit risk.

But is it right? The Fed’s accounting provides some insights into public funds investment trends over time at the macro level, but beyond this macro analysis the data is limited by sources and methods. And in some cases, it may advance misleading conclusions.  

First the good:  The bank data are derived from regulatory reports, and the mutual fund data uses benchmarks provided by the mutual fund industry that is updated annually. These are likely quite accurate.

On the other hand, the Treasury, Federal agency and corporate debt estimates derive from the 2020 Census of Governments but the allocation to security types was last calibrated in 2011.  Thus, significant changes in this segment that may have been missed.

Why it matters: The level of assets in banks is likely accurate—after all this is the central bank that compiles the report based on bank regulatory filings.  The estimate of $721 billion in bank balances as of March 31, 2023, should be considered to be “real.” It represents about 4% of total U.S. bank deposits. The money market mutual fund estimate of $71 billion is also likely accurate.  State and local governments are not major investors in money market funds. (U.S Money market fund assets totaled about $5.8 trillion at the end of the quarter.) 

Here are some conclusions from the subset of data that is likely pretty accurate:

1. state and local governments have maintained about 20% of their assets in banks during this five-year period that has seen notable changes in short-term interest rates and sharp expansion of the total of bank deposits (from about $12 trillion in 2018 to nearly $18 trillion last summer);
2. they have shifted deposits from time and savings accounts to cash and demand accounts over this multi-year period;
3. states and local governments did not run for the exit in March when the spate of bank failures hit. While corporate, institutional and retail depositors withdrew significant amounts from the nation’s banks, government deposits were down less than $10 billion versus the year-end. Perhaps this is because in many states bank deposits must be insured or collateralized for public agencies. Or perhaps the herd instinct is less pressing in the public sector.  On this theme we saw similar behavior when the financial markets froze during the Covid lockdown in February/March 2020 and investors yanked funds from money market funds. The public sector, if only because it is a minor player in money market mutual fund investments, likely did not contribute to the volatility that funds experienced.

Now to segments of the data that are less accurate: 

We’re not so sure about the estimate of Treasuries and Federal Agency holdings, though it’s probably in the right range.

The big weakness may be the estimates of commercial paper and other credit investments. Commercial paper holdings, which were estimated in the Fed’s release to be $86 billion, are likely significantly understated.  Fitch Ratings reported that LGIPs they follow held about $70 billion of commercial paper at the end of 2022. And their LGIP universe represents about 15% of state and local government investment assets that are accounted for in the in the Fed release. Participation of state and local governments in the commercial paper market is likely under-stated by the Fed. By how much? Our “guesstimate” is that public funds investors could account for 15-20% of the $1.2 trillion CP market. If so, they should be considered major players in this key market segment.

And by the way, the Fed’s Financial Accounts report ignores entirely the existence of LGIPS—in many respects the governments’ alternative to money market funds—with assets according to Fitch Ratings of at least $525 billion.  Fed analysts might say that including LGIPs raises thorny issues around double counting, but at $525 billion and growing the LGIPs should be recognized as very significant factor in asset allocation and market activities.

Good data is basic to understanding the public funds investment world, and unfortunately good data is currently lacking.