What should a central bank (not) do?

Steil, Benn

The financial crisis that began unfurling in 2008 has led to the refashioning of the model central bank governor along the lines of Churchillian war leader, willing to try anything with the money he conjures to restore economic growth. This raises important questions as to what limits, if any, elected officials should impose on such aspiring great men, and what limits markets will ultimately impose on them if elected officials forbear. This article focuses on the second of these questions.

The Pitfalls of Aggregate Demand Management

It is a fundamental postulate of today's dominant paradigm of Keynesian macroeconomics that all demand is created equal. Government spending is interchangeable with consumer spending, which is itself interchangeable with business investment. Paul Krugman (2001) illustrated the paradigm impeccably.

"The driving force behind the current [2001] slowdown is a plunge in business investment," he observed. "Over the last few years businesses spent too much on equipment and software, and ... they will be cautious about further spending until their excess capacity has been worked off," he cautioned. But "to reflate the economy," he assured us, "the Fed doesn't have to restore business investment; any kind of increase in demand will do." For as "Larry Summers says, you don't have to refill a flat tire through the puncture." How would Krugman, then, have had the Fed refill the tire punctured by plunging business investment? "Housing," he said, "which is highly sensitive to interest rates, could help lead a recovery."

But surely Krugman understood that we could have too much demand for housing, even if the economy as a whole was producing less than what he believed to be its aggregate capacity? No, he did not, so in 2002 he urged the Fed to do more. The Fed, he wrote, "needs to create a housing bubble to replace the Nasdaq bubble" (Krugman 2002). We know how that turned out. Yet today he is singing from precisely the same hymn sheet--just louder.

If Krugman and Summers could be permitted a tire analogy in support of their paradigm, perhaps I can be allowed a shower analogy in rebuttal. Imagine you get into the shower, turn on the water, and nothing comes out. You call the plumber. He tells you there's a hole in the pipes, and that it will cost you $1,000 to repair it. You tell him just to turn up the water pressure instead.

Sound sensible? Well, this is the logic behind the Fed's strategy of flooding the money pipes until credit starts flowing freely again from banks to businesses. You wouldn't expect this to work in your shower, and there's little reason to expect it to work in the commercial lending market. The credit transmission mechanism in the United States has been seriously damaged since 2007. There is a hole in the pipes. Small- and medium-sized businesses in this country are dependent on small- and medium-sized banks for access to vital credit, yet too many of these banks remain walking dead, unable to lend because their balance sheets are littered with bad commercial and real estate loans made during the boom years. And so whereas the Fed has driven its short-term lending rate down to zero, most banks will only lend on vastly greater collateral and at much higher real interest rates than before the bust. So the Fed plows on with the cheap and easy macroeconomic option: flood the pipes and see what comes out.

We've already seen the liquidity intended to boost domestic bank lending instead spill out through the cracks into markets as diverse as agricultural commodities, metals, and poor-country debt. Those bubbles will burst, as they always do, but more will be doubtlessly be created through these tried and troubled methods of modern central banking.

A Note on Liquidity Traps

When further monetary loosening flails to generate a sustainable recovery, Krugman will tell us it's because we are in a liquidity trap--a rare exception to the putative interchangeability of monetary and fiscal policy in which monetary policy becomes, in effect, chicken soup (i.e., it may not help, but it can't hurt), and only fiscal policy (specifically, increased government spending) will do.

What, then, is a liquidity trap? "The economy is in a liquidity trap when even a zero nominal interest rate isn't enough to restore full employment," explains Krugman (2011). "That's it."

This is an astonishing definition, as it implies that there is no possible reason for less than full employment when the central bank's policy rate is zero other than that the government is not spending enough. If the government were to, say, triple the minimum wage, does less than full employment at a zero policy rate still mean we're in a liquidity trap, and that the government must therefore increase spending?

What Krugman surely means is not that there is no possible reason for less than full employment at a zero policy rate, but rather that there is no permissible reason for it. I say "permissible" because he takes it as given that government intervention to prevent employment below wage, benefit, or job-security levels it finds acceptable is intrinsically good: any unemployment that might result from it can and should necessarily be offset by cheaper money and more government spending. (He cannot argue that structural unemployment is not an issue in the current liquidity trap, because his definition of liquidity trap would then be circular.)

This leads to a wider point about this paradigm. Although its exponents almost invariably want more business activities to be subject to more restrictions, and frequently fault economic downturns on the lack of such restrictions, the presence or absence of these microeconomic interventions is ultimately irrelevant to growth, in their model of the world, since growth can always be increased through macroeconomic metals--that is, by looser monetary or fiscal policy. Over time, this approach is a sure recipe for instability and stagflation.

The Limits of Monetary Sovereignty

The current crisis has been widely invoked in support of the notion that every nation, no matter how small (even a statelet like Iceland), needs its own activist national central bank. Yet it is crucial to note that no central bank outside the Federal Reserve and European Central Bank, which mint 90 percent of the world's monetary reserves, can even attempt the requisite sort of liquidity heroics. "Wow," Krugman (2011b) wrote after reading Manuel Hinds and me say as much in the Financial Times (Steil and Hinds 2011), "Have these guys ever talked to anyone in Sweden, which doesn't need euros to create more kronor?"

[FIGURE 1 OMITTED]

Well, let's look at the data--for Australia too, which Krugman throws into the mix. As Figure 1 shows clearly, when the Swedish and Australian central banks expanded credit dramatically during the recent financial crisis they also liquidated foreign assets. (1) That is, both central banks accompanied their massive easing with a scramble for reserve currencies. They were unable merely to conjure their own monetary resources for the task. And this is not merely an artifact of the most recent crisis, as three full decades of Australian data show. Any country that ploughs on creating credit without ample reserve currencies (dollars and euros) eventually becomes a ward of the IMF.

The Perils of Delegating Fiscal Authority to Central Banks

The traditional argument for central bank independence was based on the notion that monetary policy was a uniquely important aspect of economic policy and that its conduct needed to be isolated from short-term political pressures. The ECB and the Fed have, however, since 2008 arrogated powers to create and allocate credit that greatly exceed those that legislators ever intended to grant them. This is typically defended on the grounds that the political process is too slow and cumbersome to deal with the scale, or potential scale, of the problems created by a rapid and systematic withdrawal of credit to systemically important private borrowers or to sovereigns. The logical possibility that tills is so can hardly be denied, but neither can the democratic reasons for having strictly limited such powers in the first place. As compelling as such reasons are, I put them aside here, and make only the narrow point that the ECB, in particular, still faces hard limits on its ability to conduct monetary policy while plugging gaps for reticent fiscal authorities.

Back in 2000, the ECB's first president, Wim Duisenberg, explained that he knew the Bank's operational framework for implementing monetary policy was working well because it was successfully steering short-term market interest rates exactly where the Bank wanted them to go. Prior to the crisis, the ECB's policy rate was indeed tightly connected to 3-month eurozone government borrowing rates. In a growing swath of the eurozone, however, this relationship has collapsed, as shown in Figure 2--in the case of Spain, from nearly 100 percent to nearly 0 percent. Default risk increasingly dominates these rates, which themselves strongly influence rates in the private sector. By Duisenberg's criterion, monetary policy in the eurozone is becoming less and less effective as the ECB wades deeper and deeper into the political quicksand of fiscal policy and discretionary credit allocation.

In continually insisting that a eurozone sovereign debt restructuring was out of the question, the ECB was merely talking its book; that is, trying to avoid the threat to its independence that would follow if the Bank were itself to need a recapitalization by the member states.

That day is coming perilously close. The ECB has bought over 200 billion [euro] in bonds of the riskiest eurozone sovereigns since May 2010, and has loaned over three times that much to eurozone banks, often against highly dubious collateral. With only 81 billion [euro] in capital, a mere 25 percent haircut on its direct and indirect Greek, Irish, and Portuguese bond holdings would render it technically insolvent. Without sufficient assets to sell, it would have no effective means of withdrawing excess euros from circulation in the future. A timely recapitalization, which would have to be largely undertaken by the German taxpayer, cannot be taken as a given. The imminent prospect of a technically insolvent ECB could therefore precipitate a major run on the euro.

[FIGURE 2 OMITTED]

The Fed is thankfully in a much less precarious position. Though leveraged 56:1, and post-"Operation Twist" could see its reported capital wiped out with a roughly 33 basis point rise in long-term yields, profits on its securities purchases have to date provided it with ample cushion ($78 billion in 2010, vs. $58 billion in reported capital). Its balance sheet is also credibly backstopped by its national Treasury, in contrast to the ECB. The Fed's $1 trillion mortgage security portfolio is, however, politically toxic, in the sense that it cannot be readily liquidated to tighten monetary policy without provoking a political firestorm, not to mention a possible cratering of the government-dependent U.S. housing market. The only remedy for this serious threat to long-term monetary stability is to hand the portfolio over to the Treasury in exchange for Treasury securities, which would normalize the composition (though certainly not the size) of the Fed's balance sheet. In the end, this amounts to placing the liability for this extraordinary housing-market intervention with the Treasury, where it should have been along. The central bank's balance sheet is ultimately a dangerous substitute for political will and public accountability.

References

Krugman, P. (2001) "Reckonings; Delusions of Prosperity." New York Times (14 August).

--(2002) "Dubya's Douple Dip?" New York Times (2 August).

--(2011a) "Liquidity Traps, Once Again." New York Times (18 March).

--(2011b) "Reserve Currency Mysticism." New York Times (29 May).

Steil, B., and Hinds, M. (2011) "Keynesians Are Complacent about the Dollar." Financial Times (23 May).

(1) I am indebted to Manuel Hinds for these data.

Benn Steil is Director of International Economics and a Senior Fellow at the Council on Foreign Relations.