Will the Fed go bankrupt?

The simple answer is no. A few years ago, the Fed quietly issued a memorandum in which it stated that any bond losses the agency suffered would be made up by the Treasury (thus by the taxpayers). But that doesn’t mean that the Fed can’t encounter serious losses on its bond holdings, right? Well, new research suggests otherwise.

USD 3.3 trillion

As shown in the graph below, the Fed increased its balance sheet by a mind blowing USD 3.3 trillion since the outbreak of the financial crisis in 2008. Most of this historical increase is related to purchases of Treasuries and mortgage securities that are backed by the government. Hence, the risk the Fed bears on its balance sheet is predominantly related to changes in interest rates. Credit risk is mostly absent.

Click to enlarge


So, what if interest rates normalize? Should the Fed expect big losses on its portfolio? A recent study by Christensen, Lopez and Rudebusch (2014), linked to the Federal Reserve Bank(!) of San Francisco, suggests that the chance of mark-to-market losses on the Fed bond portfolio as a result of higher interest rates is actually pretty slim.

Interest Rate Scenarios

Christensen, Lopez and Rudebusch generate a range of interest rate scenarios by taking both historical data as well as the constraint that nominal interest rates cannot go below zero into account. This way they try to obtain a realistic series of (asymmetric) interest rate forecasts which are then used to determine the impact on the value of the Fed bond portfolio.

The graph below shows their main findings. The blue line marked 50th percentile, ‘represents the median portfolio value, which is expected to decline over time as bonds mature and as interest rates rise from their current historic lows. The future value of the Fed’s Treasury portfolio is equally likely to be above or below these median values.’

Click to enlarge


Low probability of Fed losses

The other lines represent scenarios, with different probabilities, in which interest rates rise. As the graph reveals, even in an interest rate scenario which has only a probability of 5% (the red 5th percentile line), the projected market value of the Fed’s Treasury securities will not fall below the face value of the portfolio. Only in scenarios with lower than 5% probability the Fed is expected to encounter losses on its bond assets. Or, in other words, only unusual (low probability) increases of interest rates would leave the Fed with losses on its portfolio. According to this research, The Fed does not have to worry all that much about rising interest rates when looking at its balance sheet.

I would like to end with two short remarks. First, the study uses data until the end of 2012, while the Fed has kept buying ever since. However, Christensen, Lopez and Rudebusch state that results are pretty similar when using data until 2013. This sounds credible as the authors focus primarily on the probability of scenarios in which rates go up so significantly that the market value of the bond portfolio decreases below its face value. Bigger numbers do not necessarily lead to different probabilities. But, the potential size of the loss, measured in US dollars, certainly does change with a bigger balance sheet. So, the dollar amount that the government (or the taxpayer) potentially has to cough up is in fact very much related to the size of the balance sheet. Second, the authors do not provide an indication on what kind of rate rise is accompanied with the low probability scenarios they present. This would, however, make the scenarios more tangible and also more comparable to previous studies that did focus on the size of the rate increase.


Christensen, Jens H.E., Jose A. Lopez, and Glenn D. Rudebusch. 2014. “Stress Testing the Fed.” FRB San Francisco Economic Letter 2014-08. http://www.frbsf.org/economic-research/publications/economic-letter/2014/march/federal-reserve-interest-rate-risk-stress-test/

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s