VoxEU Column Financial Markets Monetary Policy

Financial stability and monetary policy

The monetary policies implemented by the Federal Reserve since late 2008 have raised concerns about the risk taking of financial institutions. This column discusses the effect of some of these policies on life insurance companies and market mutual funds. While the effect on life insurance companies has been stabilising, money market funds did not actively reach for yield.

In the winter of 2008, the Federal Reserve began an unprecedented campaign to combat the economic downturn. The mix of policy instruments included a near zero federal funds rate, explicit communication regarding the forward path of the funds rate, and a balance sheet that ballooned to more than $4 trillion as of this writing. With memories of the 2008-09 financial crisis still fresh, the policies have prompted concern for their effect on financial stability (Bernanke 2013, Stein 2013, Fisher 2014, Yellen 2014).

Analysis of the possible trade-offs of financial stability requires understanding of how monetary policy affects financial stability. Ceteris paribus, lowering the safe interest rate makes riskier investments relatively more attractive. As emphasised recently by the Bank of England's chief economist (Financial Times 2014), the resulting increase in risk taking constitutes an intentional effect of monetary policy. Risk taking might further increase if some asset managers ‘reach for yield’ because of some principle-agent friction or institutional aversion to low returns, possibly pushing risk premia below their first best level (Rajan 2005). On the other hand, many of the financial institutions at the heart of the 2008-09 panic already hold substantial quantities of risky assets on their balance sheets. By aiding the real economy, monetary policy helps to reduce loan default rates and raises firm profits, resulting in improved performance and higher values of these legacy assets. This ‘stealth recapitalisaton’ may reduce the riskiness of financial institutions even if those institutions simultaneously increase their risk taking. Similarly, many financial institutions sell products with positive income elasticities, such that new business recovers along with the broader economy.

In a recent paper (Chodorow-Reich 2014), I assess empirically the effects of easy monetary policy on various financial institutions. Life insurance companies and money market mutual funds provide good illustrations of the channels just described.

Life insurance companies

The combination of long-term fixed income liabilities and shorter duration assets makes life insurers vulnerable to unexpected declines in interest rates. The prospect of a diminished or even negative interest spread as bonds get rolled over at lower interest rates can cause reaching for yield in effort to avoid the decline in net income. Thus, low interest rates could prompt additional risk taking by life insurance companies.

Conversely, many life insurers faced solvency crises in early 2009 because of large write-downs on their holdings of corporate bonds and mortgage-backed securities. These legacy assets stood to benefit from an improving real economy.

I use high-frequency event studies to measure the response of credit default swap (CDS) spreads, bond yields, and equity prices of life insurers in narrow windows surrounding surprise announcements of policy changes. Figure 1 displays the equity response of different life insurers (or their parent companies) during thirty minute windows surrounding the announcements (results for bond prices and CDS spreads are similar). The figure shows a scatterplot of the stock price change and the change in the five year Treasury during the announcement window, with the announcement date labelled on the lower horizontal axis.

Figure 1. Insurance company stock price response to unconventional monetary policy surprises

Notes: The log change in stock price is from two minutes before to 18 minutes after the announcement. The change in the on-the-run five-year Treasury is the change in the yield to maturity during the same window.

Two important patterns emerge:

  • First, consistent with large effects on the value of the legacy asset holdings, the introduction of near zero interest rates and quantitative easing in the winter of 2008-09 led to sharp increases in life insurers' stock prices.

Indeed, every life insurer in the sample experienced an increase in its stock price during the windows surrounding the 16 December 2008 announcement of a 75-basis-point reduction in the federal funds rate to a new target of 0-25 basis points, and the 18 March 2009 balance sheet expansion of up to $1.15 trillion.

  • Second, while policy announcements since early 2009 have had smaller effects on both Treasury prices and the stock prices of life insurers, the sign has not changed.

For example, when Chairman Bernanke set off the ‘taper tantrum’ of summer 2013 by suggesting during Congressional testimony in May of that year that the Federal Open Market Committee could “take a step down in the pace of [asset] purchases” during coming meetings, 12 of the 13 life insurers in the sample experienced an immediate decline in their stock price.

Money market mutual funds

The effect of low interest rates on reaching for yield by money market funds has also generated concern. Near zero interest rates on short-term liquid assets squeeze funds' ability to cover their administrative costs. Because investors can always put their money into zero interest checking accounts, money market funds have responded to the low rate environment by waiving substantial portions of the fees normally charged to investors. Essentially all funds have waived exactly enough of their fees to leave investors with an after-fee return of between zero and five basis points.

The necessity of leaving investors with a weakly positive after-fee return creates a possible incentive for funds to reach for higher returns to avoid waiving fees. In fact, beginning in 2009, funds passed higher gross yields through into higher charged expenses nearly one-for-one. If additional risk taking causes some of a fund's securities to lose even a small amount of value, the fund may have to ‘break the buck’, causing a broad run on money market funds similar to what ensued following the Reserve Primary Fund's breaking the buck in September 2008.

To assess whether the desire to avoid fee waivers may have pushed funds to increase their risk taking, I exploit cross-sectional differences in fund administrative costs. In 2006, for example, the 10th percentile fund had overhead costs of 27 basis points, while the 90th percentile fund had costs of 103 basis points. If funds sought to avoid operating losses, then we would see the higher administrative cost funds moving into relatively riskier and higher yield assets in the low interest rate environment. In this vein, Figure 2 plots the coefficients from a weekly instrumental variables regression. The dependent variable is the average yield on the securities in a fund's portfolio, and thus measures whether a fund has reached for higher yield. The plotted coefficients give the increase in the gross yield corresponding to a percentage point increase in the ratio of the fund's overhead costs to assets, with the ratio instrumented using its 2005 value. The regression also controls for fund fixed effects and the fund's 2005 asset allocation, with the fixed effect separately identified by including observations from 2006 for all regressions.

Figure 2. Loading of gross yield on incurred expenses, by week

Notes: The solid line plots the coefficients from a weekly regression of gross yield on incurred expenses, with incurred expenses instrumented using their 2005 average value. The regression also contains fund fixed effects identified by including observations from 2006, and fund type and asset class holdings in 2005 interacted with week. Sample excludes fund-weeks with a gross yield of zero. The dotted lines plot 95% confidence interval bands based on standard errors clustered by fund sponsor.

The figure shows high cost funds pursuing higher gross returns in 2009-11. The first jump in the coefficient occurs in October 2008 – the first month in which the average yield on a one-year Treasury bill fell below 50 basis points. Even at the peak, however, the economic magnitude is small; a coefficient of 0.2 implies that a one percentage point (or roughly three standard deviations) increase in the ratio of overhead costs to assets leads to an additional 20 basis points in yield. Notably, such behaviour vanishes by 2013. Perhaps paradoxically, the low volatility environment induced by low interest rates had the effect of compressing yields across asset classes, leaving little room for funds to reach for yield.

Conclusion

In sum, the policies pursued by the Federal Reserve since late 2008 have affected financial institutions’ risk taking. Strikingly, the policies' cumulative effect on life insurance companies appears to have been stabilising, as the benefit to legacy asset values and demand for new products outweighed any reaching for yield. While some money market funds did engage in reaching for yield in 2009-11, the compression in spreads in recent months has sharply limited the scope for such behaviour. I report similar results for bank holding companies and private pension funds, respectively. At this juncture, financial-stability concerns for monetary policy should not stem from concerns about the riskiness of these sectors.

References

Bernanke, B (2013), “Testimony of Chairman Ben S. Bernanke before the Joint Economic Committee, U.S. Congress,” Washington, DC, May 22, 2013.

Chodorow-Reich, G (2014), “Effects of Unconventional Monetary Policy on Financial Institutions.” Forthcoming in Brookings Papers on Economic Activity.

Fisher, R (2014), “Beer Goggles, Monetary Camels, the Eye of the Needle and the First Law of Holes.” Remarks presented to the National Association of Corporate Directors, January 14, Dallas.

Gilles C and S O’Connor (2014), “Haldane backs central banks’ actions in a ‘nutty’ world”, Financial Times, 2 July.

Rajan, R (2005), “Has Financial Development Made the World Riskier?” In Proceedings of the Jackson Hole Conference. Kansas City: Federal Reserve Bank of Kansas City.

Stein, J (2013), “Overheating in Credit Markets: Origins, Measurement, and Policy Responses.” Remarks presented at the research symposium, “Restoring Household Financial Stability after the Great Recession: Why Household Balance Sheets Matter,” sponsored by the Federal Reserve Bank of St. Louis, St. Louis, Missouri, February 7.

Yellen, J (2014), “Monetary Policy and Financial Stability” Michael Camdessus Central Banking Lecture. Washington, DC: International Monetary Fund.

3,465 Reads