Isn’t disclosure always a good thing?  Isn’t it always better to have audited financial statements than not (ignoring the cost of the audit itself)? Wouldn’t it be particularly important to audit a large bank?  How about one that holds our entire financial system in its hands…such as the Federal Reserve Bank?

I was very surprised to read GMU economist Tyler Cowen claim that “The Fed has resisted being audited“, and related posts by other GMU-affiliated faculty debating the issue. (See here, for example).

A little searching suggests that economists and accountants don’t speak the same language or attend to the same issues.  In fact, the Fed does provide audited financial statements, which you can find here.  The dispute really seems to be about whether the GAO should be examining and reporting on the Fed’s policy and operational decisions and processes, as indicated in Bernanke’s letter in May.  That may or may not be a good idea, but it isn’t really what comes to my mind when I hear the words “Audit the Fed.”

However, the searching was still worthwhile, as I came across a very interesting paragraph regarding the Fed’s financial statements.  It turns out, the Fed doesn’t use GAAP, but instead uses accounting specific to Federal Reserve Banks.  As they explain (all emphasis mine):

The primary difference between the accounting principles and practices in the Financial Accounting Manual and GAAP is the presentation of all System Open Market Account (SOMA) securities holdings at amortized cost rather than the fair value presentation required by GAAP. Treasury securities, government-sponsored enterprise (GSE) debt securities, Federal agency and GSE mortgage-backed securities, and investments denominated in foreign currencies comprising the SOMA are recorded at cost on a settlement-date basis rather than the trade-date basis required by GAAP. The cost basis of Treasury securities, GSE debt securities, and foreign government debt instruments is adjusted for amortization of premiums or accretion of discounts on a straight-line basis. Amortized cost more appropriately reflects the Reserve Banks’ securities holdings given the System’s unique responsibility to conduct monetary policy. Accounting for these securities on a settlement-date basis more appropriately reflects the timing of the transaction’s effect on the quantity of reserves in the banking system. Although the application of fair value measurements to the securities holdings may result in values substantially above or below their carrying values, these unrealized changes in value have no direct effect on the quantity of reserves available to the banking system or on the prospects for future Reserve Bank earnings or capital. Both the domestic and foreign components of the SOMA portfolio may involve transactions that result in gains or losses when holdings are sold prior to maturity. Decisions regarding securities and foreign currency transactions, including their purchase and sale, are motivated by monetary policy objectives rather than profit. Accordingly, fair values, earnings, and gains or losses resulting from the sale of such securities and currencies are incidental to the open market operations and do not motivate decisions related to policy or open market activities.

I haven’t yet done enough searching to determine whether the fair values are disclosed, even if they aren’t recognized, and what the quality of that disclosure is.  But it is worth pointing out that audits and disclosure are not always wonderful things.  Disclosure can reduce welfare by exposing everyone to the risk of price changes (when the bad news comes out), as discussed in Dye’s 2001 JAE discussion paper.  In the case of systemic risk, disclosure of bad news could trigger a bank run, making society better off with worse disclosure.  I am not saying the Fed shouldn’t disclosure its financial position (including fair values)…just that the questions at the start of this post don’t have answers as obvious as you might have thought.