Shadow banking and the economy

Alan Moreira, Alexi Savov 16 September 2014

a

A

Shadow banking, what is it good for? At the epicentre of the global financial crisis, shadow banking has become the focus of intense regulatory scrutiny. All reform proposals implicitly take a stance on its economic value.

According to the prevailing regulatory arbitrage and neglected risks views, it doesn’t have any – shadow banking is about evading capital requirements, exploiting ‘too big to fail’, and marketing risky securities as safe to unwitting investors. The right response is to bring shadow banking into the regulatory and supervisory regime that covers insured banks.

A third view, the liquidity transformation view, is different. It says that shadow banking adds value by creating money-like liquid securities from risky illiquid assets (Gorton and Metrick 2010). While this process is inherently unstable, simply shutting it down risks damaging liquidity provision to the real economy.

The liquidity transformation view confronts us with a real tradeoff between financial stability and economic growth. In a recent paper (Moreira and Savov 2014) we formalise this tradeoff within a dynamic macroeconomic model. In this column we explain our framework and draw its implications for regulatory reform.

People want money

The last two decades have seen a dramatic rise in worldwide demand for safe and liquid securities, or money for short.1 These securities enable the large volume of transactions that course through the modern economy. Yet the supply of truly safe (typically government-backed) assets needed to back them has not kept up. Shadow banking has been making up the difference via money market funds, asset-backed commercial paper, and dealer repurchase agreements.2

The key feature of these instruments is that they function just as well as traditional money (think of bank deposits or short-term government obligations) during quiet times, but they abruptly lose their liquidity at the first sign of trouble.3 Of course, it would be better to have fully stable liquidity, but in a world of limited good collateral, shadow banking may well be the next-best alternative.

Shadow banking booms and busts

In our paper, we show that the liquidity expansion enabled by shadow banking leads to a lower cost of capital for firms, greater investment, and a higher level of economic growth. Moreover, during a shadow banking boom, the economy moves up the risk-return frontier, funding riskier but more productive investments. Over time, this process builds up fragility.

At the peak of a shadow banking boom, even a modest shock can set off a cascade of adverse events. Faced with increased uncertainty, investors are no longer willing to hold shadow banking securities whose liquidity could easily evaporate. Shadow banking shuts down. The flight to safety forces the financial sector to de-lever as liquidity creation now requires more collateral. The liquidity crunch forces discount rates up and asset prices down. The instability of prices drives up margins, leading to a downward spiral that exacerbates the collateral shortage and the liquidity crunch. The end result is that capital investment falls and the economy enters a deep recession.

The problem with shadow banking then is that while it makes the good times better, it also makes the bad times worse. In the language of economics, shadow banking imposes a negative externality on the rest of the economy. During the boom, individual institutions have little incentive to take into account the impact of their actions on the severity of the bust. Shadow banking allows them to fund riskier assets more cheaply, but doing so decreases the economy’s supply of good collateral when times get rough.

Implications for policy

This collateral externality represents a sound rationale for government intervention. Our work shows that the precise nature of the intervention is important, as it needs to strike the right balance between curtailing excessive risk-taking and preserving liquidity provision.

At the heart of a shadow banking bust is an acute shortage of good collateral resulting from the risky investments made during the boom and the margin spirals that unfold in the bust. The central bank can alleviate this shortage by buying up risky assets and providing safe ones, as the Fed did in the autumn of 2008. A well-timed intervention can arrest the financial collapse and stabilise the economy.

Interestingly, a maturity extension programme like ‘Operation Twist’ can be counterproductive. The reason is that long-duration safe assets like Treasury bonds tend to appreciate in a crisis. This means that they serve as a hedge when combined with risky assets on bank balance sheets. Treasury bills and excess reserves do not have this property. This type of programme cannot be judged by its effect on long-term Treasury yields, as it tends to lower them even as it is making the financial sector more constrained, not less.

More recently, the Fed has been implementing a reverse repurchase programme that allows it to provide liquidity directly to non-banks. The Fed sells bonds from its balance sheet to money market funds in exchange for cash (excess reserves), then repurchases the bonds on the following day with a bit of interest. Until now, this process has taken place in the private repo market.

The attraction of the reverse repo programme is that it provides a stable source of liquidity to a wide variety of institutions. At the same time, it has important limitations that have not received attention. The Fed’s bond holdings are very safe, which means that private intermediaries are left holding riskier ones. As a result, reverse repo might crowd out fully safe private intermediation like Treasury-backed repo, and actually crowd in riskier shadow banking activity.

A reverse repo programme backed by lower-quality collateral (or no collateral at all) would not have this downside, though it could expose the Fed to losses. Such a programme would be closer to the realm of fiscal policy, and might best be left to the Treasury. For example, the Treasury could issue floating-rate debt. The benefits of doing so should be weighed against distortions arising from increased taxation.

The bottom line

Above all, our point is that when it comes to reforming the shadow banking system, there are serious tradeoffs. Six years after the collapse of Lehman, these tradeoffs need to be understood so that we can grapple with the question: By how much should we lower the height of the boom to ensure a softer landing in the bust?

References

Gorton, G and A Metrick (2010), “Haircuts”, Federal Reserve Bank of St Louis Review, 92(6): 507–519.

Kacperczyk, M and P Schnabl (2013), “How Safe are Money Market Funds?”, Quarterly Journal of Economics, 128(3): 1073–1122.

Moreira, A and A Savov (2014), “The Macroeconomics of Shadow Banking”, Working paper. 

Pozsar, Z (2014), “Shadow Banking: The Money View”, Office of Financial Research Working Paper 14-04. 

Sunderam, A (2012), “Money creation and the shadow banking system”, mimeo, Harvard Business School. 

Footnotes

1 Pozsar (2014) shows that large institutional cash pools have grown in size from $2 trillion in 1997 to over $5.5 trillion today. The sources of this increase include central banks, corporations, and asset managers.

2 Sunderam (2012) finds that issuance of asset-backed commercial paper responds to the premium for money-like Treasury bills.

3 Kacperczyk and Schnabl (2013) show that the spreads on various instruments held by money market funds widened early on in July 2007 following the failure of two Bear Stearns hedge funds.

a

A

Topics:  Financial markets Global crisis Macroeconomic policy

Tags:  shadow banking, banking, financial crisis, global crisis, regulatory arbitrage, liquidity transformation, financial stability, externalities, collateral, business cycle, financial regulation, financial fragility, liquidity, liquidity crunch

Assistant Professor of Finance in the School of Management, Yale University

Assistant Professor of Finance at the Stern School of Business, New York University

Events