A view that is gaining popularity as one of the fundamental explanations for the current crisis is that emerging markets’ voracious appetite for international reserves coupled with record-low US policy interest rates and lax financial regulation to produce a frantic “search for yield,” the creation of fragile financial instruments, and occasionally outright fraud. For example see Henry Paulson’s discussion quoted in Guta (2009).
This view – particularly, the “financial fragility” component – could help to answer a central question, namely, why minor fireworks in the subprime mortgage market ignited a fearsome powder keg and a local problem became global in a short span of time.
In this column, I will present a framework that provides some conceptual support for the view. The framework stresses fragilities associated with liquid financial instruments that have long been identified in the finance literature.1 For the sake of concreteness, I will focus on the Fed and abstract from international aspects, unless strictly necessary.
The financial framework
The argument develops through eight related points:
1. A starting point is that the 1997/8 Asian/Russian crises showed emerging economies the advantage of holding a large stock of international reserves to protect their domestic financial system without IMF cooperation. This self-insurance motive is supported by recent empirical research, though starting in 2002 emerging economies’ reserve accumulation appears to be triggered by other factors.2 I suspect that a prominent factor was fear of currency appreciation due to: (a) the Fed’s easy-money policy following the dot-com crisis, and (b) the sense that the self-insurance motive had run its course, which could result in a major dollar devaluation vis-à-vis emerging economies’ currencies.3
2. Let me make some simplifications. I will assume that reserve money is a composite of US currency and Treasury bills. Let s be the nominal interest rate on reserve money.4 Thus, when the demand for international reserves goes up, the Fed can opt for accommodating its supply or lowering the policy interest rate (which I will equate with s).
3. Enter the private sector as producer of reserve money and, as I will conjecture, generator of a rickety financial system. Asset-backed securities and collateralised debt obligations are different from Treasury bills but are certainly much closer to reserve money than the underlying assets. Thus, the development of those instruments can be seen as helping to create what might be called (reserve) quasi-money.
Quasi-money creation is costly; part of the cost stems from the fact that quasi-money competes with official reserve money. When s declines – especially when s falls more than inflation as in the US – the marginal cost of creating quasi-money goes down, stimulating supply. Therefore, an increase in the demand for international reserves accompanied by a lower interest rate on reserve money (s) will give rise to an increase in the supply of quasi-money. The effect of low s is enhanced by lax financial regulation and the expectation of bailouts in case of systemic crisis (more on this below). Without the latter, the supply effect was unlikely to be large.
4. As a general rule, quasi-money can be created by generating some type of mismatch of maturities or currency denomination. For example, bank deposits are a class of quasi-money which has shorter maturity than the assets banks hold against them. Therefore, their moneyness requires that only a handful of depositors attempt to cash their deposits at the same time. If rumour spreads that depositors will massively try to withdraw their deposits, depositors will have strong incentives to do the same, which results in widespread bank failures, destroying the moneyness of deposits. This has been one of the central motivations for the creation of central banks.5
5. We now know that the new financial instruments were partially insured by regular banks through, for example, structured investment vehicles. Learning about that seems to have startled many observers and regulators who thought that securitisation had taken meltdown risks off of banks’ balance sheets. However, a little thinking should have warned them that such risk transfer was bound to be incomplete, because banks can piggy back on central banks, especially in a systemic crisis, as actually happened.6
6. Under these circumstances, banks would be called to honour the insurance contracts if a run against quasi-money materialises, thus forcing central banks to come to their rescue.
Unfortunately, given the nature of their mandates, central banks stepped in only when regular banks were on the verge of collapse because insurance arrangements had been activated and they did not have the resources to meet them. At that juncture, the quasi-money’s credibility had already been lost and the financial system was stuck in a situation in which the supply of quasi-money had correspondingly collapsed.
Summary of points 1 to 6
To summarise, the increase in the demand for international reserves, accompanied by low US policy interest rates and lax financial regulation, may have led to a large-scale creation of quasi-money subject to self-fulfilling-expectations runs. The probability of runs against the new instruments was presumably low but likely much higher than for bank deposits. Central banks eventually reached the source of the financial problems but damage to the credibility of the financial sector had already occurred. Liquidity collapsed, setting in motion strong price-deflation forces.
Real sector impact
Let’s turn to the non-financial or real sector.
7. Keeping banks and other institutions afloat does not guarantee that credit will be revived and that credit flows will go back to normal. There are three independent reasons for credit flows to dry up.
- First, prior to crisis, credit flows were partially structured on instruments that are no longer available or have drastically lost their appeal.
- Second, price deflation could give rise to Irving Fisher’s debt deflation and widespread bankruptcy.7
- Third, part of the stock of quasi-money was based on asset-backed securities; as their moneyness evaporates, the relative price of the underlying assets (e.g., real estate) falls, lowering available collateral and, consequently, further dampening credit.8
8. A sudden stop of credit flows has a direct impact on the real sector9, forcing a sudden and large cut in private sector expenditure (a flow).10 In particular, large cuts in the flow of credit for working capital results in sizable falls in investment and employment. Moreover, since it is unlikely that expenditure contraction will be uniform across the economy, the credit sudden-stop may give rise to sharp changes in relative prices, further complicating the financial landscape. Bad debts will arise but they may be just a consequence of quasi-money destruction, not of over-borrowing.
There are six key policy implications:
1. Financial innovation and bubbles could stem from lax monetary policy and financial regulation.
2. Bubbles are not all the same. Bubbles that involve the banking system are likely the worst kind, because they could bring about a sudden stop of bank credit, seriously draining working capital, for example.
3. With the benefit of hindsight, to prevent price deflation in the first half of the 2000s, the Fed should have resorted to quantitative easing instead of keeping interest rates low for an extended period of time. This would have signified a radical departure from the Fed’s practice and, in all probability, would have been difficult to defend or even explain in a no-deep-crisis environment.
Going forward, however, the Fed (or whichever its successor may be) should add quantitative easing to its tool kit in normal situations and employ it to accommodate a major increase in the demand for reserve money. To operationalise this, the Fed could, for example, have a rule by which quantitative easing is triggered once its policy interest rate reaches a lower bound, larger than zero. For example, the lower bound could be made equal to the long-run marginal productivity of capital plus target inflation.
4. During financial crises, expansive monetary and fiscal policy may not suffice. An aggressive credit policy may be called for. Since under those circumstances credit markets don’t work properly, the central bank may have to direct credit to strategic sectors, like Brazil has done on several occasions.
5. Crisis time is no time for implementing tighter financial regulation. The latter may exacerbate contraction of credit flows and enhance its deleterious effects.
6. The above observation weakens any tough statement in normal times about policy in crisis times (e.g., a commitment to no-bailout). But, normal times are the time to deactivate financial bombs.
The main challenge is that the financial sector is in constant evolution, and regulators are required to be “ahead of the curve.” Thus, it would be advisable for the regulatory authority to have a unit closely following developments in the capital market. Given globalisation, this task should be coordinated with other regulatory authorities. The BIS and the IMF could play a key role in this respect.
1. See Allen and Gale (2007). See Calvo (2009a, 2009b) for models that highlight the macroeconomic role of liquid instruments.
3. Sometimes this policy is called “neo-mercantilism.” However, emerging markets’ intervention in the foreign-exchange market could also be interpreted as a defensive move vis-à-vis the US beggar-thy-neighbor policy implied by its lax monetary stance.
5. See Allen and Gale (2007) for a discussion of this and other related issues.
6. This applies to the US. In emerging markets, the ability of central banks to operate as lenders of last resort in terms of reserve money depends on external credit lines and their stock of international reserves. This, by the way, is one of the reasons for the self-insurance motive.
7. See Fisher (1933). For a modern discussion of debt deflation in the context of the Great Deflation, see Bernanke (2000). For the relevance of this concept for emerging market crises, see Calvo (2005).
8. See Calvo (2009a, 2009b) for models in which the relative price of quasi-money real underlying assets falls as quasi-money liquidity evaporates
9. In line with the sudden-stop literature for emerging markets, I define a sudden stop of domestic credit as a fall in credit flows to the private sector that exceeds two standard deviations; the latter is computed on the basis of the credit-flow time series prior to each point in time. For more details, see Calvo (2009b).
10. Notice that I am referring to flows, not stocks. Stocks may not decline and still a fall in credit flows may have major real effects. This is fully in line with the literature on sudden stops of international capital inflows. See Calvo (2005).
Allen, Franklin, and Douglas Gale (2007), Understanding Financial Crises, New York, NY: Oxford University Press.
Bernanke, Ben (2000), Essays on the Great Depression, Princeton, NJ: Princeton University Press.
Calvo, Guillermo (2005), Emerging Markets in Turmoil: Bad Luck or Bad Policy? Cambridge, MA: MIT Press.
Calvo, Guillermo (2009a), “Financial Crises and Liquidity Shocks: A Bank-Run Perspective,” NBER Working Paper 15425.
Calvo, Guillermo, (2009b), “Looking at Financial Crises in the Eye: A Simple Finance/Macro Framework” Columbia University mimeograph.
Calvo, Guillermo, and Carlos Vegh (1995), “Fighting Inflation with High Interest Rates: The Small-Open-Economy under Flexible Prices,” Journal of Money, Credit, and Banking, 27, pp 49-66.
Canzoneri, Matthew, Robert E. Cumby, Bezhad Diba, and David Lopez-Salido (2008), “Monetary Aggregates and Liquidity in a Neo-Wicksellian Framework,” Journal of Money, Credit, and Banking, 40, 8, December, pp. 1667-1698.
Fisher, Irving (1933), “The Debt-Deflation Theory of Great Depressions,” Econometrica, pp. 337-357.
Guha, Krishna (2009). “Paulson Says Crisis Sown by Imbalance.” Financial Times, 1 January.
Obstfeld, Maurice, Jay C. Shambaugh, and Alan M. Taylor (2008), “Financial Stability, the Trilemma, and International Reserves,” NBER Working Paper 14217.