venerdì 1 marzo 2013
Ten QE Questions, by Nouriel Roubini
Most observers regard unconventional monetary policies such as quantitative easing (QE) as necessary to jump-start growth in today’s anemic economies. But questions about the effectiveness and risks of QE have begun to multiply as well. In particular, ten potential costs associated with such policies merit attention.
First, while a purely “Austrian” response (that is, austerity) to bursting asset and credit bubbles may lead to a depression, QE policies that postpone the necessary private- and public-sector deleveraging for too long may create an army of zombies: zombie financial institutions, zombie households and firms, and, in the end, zombie governments. So, somewhere between the Austrian and Keynesian extremes, QE needs to be phased out over time.
Second, repeated QE may become ineffective over time as the channels of transmission to real economic activity become clogged. The bond channel doesn’t work when bond yields are already low; and the credit channel doesn’t work when banks hoard liquidity and velocity collapses. Indeed, those who can borrow (high-grade firms and prime households) don’t want or need to, while those who need to – highly leveraged firms and non-prime households – can’t, owing to the credit crunch.
Moreover, the stock-market channel leading to asset reflation following QE works only in the short run if growth fails to recover. And the reduction in real interest rates via a rise in expected inflation when open-ended QE is implemented risks eventually stoking inflation expectations.
Third, the foreign-exchange channel of QE transmission – the currency weakening implied by monetary easing – is ineffective if several major central banks pursue QE at the same time. When that happens, QE becomes a zero-sum game, because not all currencies can fall, and not all trade balances can improve, simultaneously. The outcome, then, is “QE wars” as proxies for “currency wars.”
Fourth, QE in advanced economies leads to excessive capital flows to emerging markets, which face a difficult policy challenge. Sterilized foreign-exchange intervention keeps domestic interest rates high and feeds the inflows. But unsterilized intervention and/or reducing domestic interest rates creates excessive liquidity that can feed domestic inflation and/or asset and credit bubbles.
At the same time, forgoing intervention and allowing the currency to appreciate erodes external competitiveness, leading to dangerous external deficits. Yet imposing capital controls on inflows is difficult and sometimes leaky. Macroprudential controls on credit growth are useful, but sometimes ineffective in stopping asset bubbles when low interest rates continue to underpin generous liquidity conditions.
Fifth, persistent QE can lead to asset bubbles both where it is implemented and in countries where it spills over. Such bubbles can occur in equity markets, housing markets (Hong Kong, Singapore), commodity markets, bond markets (with talk of a bubble increasing in the United States, Germany, the United Kingdom, and Japan), and credit markets (where spreads in some emerging markets, and on high-yield and high-grade corporate debt, are narrowing excessively).
Although QE may be justified by weak economic and growth fundamentals, keeping rates too low for too long can eventually feed such bubbles. That is what happened in 2000-2006, when the US Federal Reserve aggressively cut the federal funds rate to 1% during the 2001 recession and subsequent weak recovery and then kept rates down, thus fueling credit/housing/subprime bubbles.
Sixth, QE can create moral-hazard problems by weakening governments’ incentive to pursue needed economic reforms. It may also delay needed fiscal austerity if large deficits are monetized, and, by keeping rates too low, prevent the market from imposing discipline.
Seventh, exiting QE is tricky. If exit occurs too slowly and too late, inflation and/or asset/credit bubbles could result. Also, if exit occurs by selling the long-term assets purchased during QE, a sharp increase in interest rates might choke off recovery, resulting in large financial losses for holders of long-term bonds. And, if the exit occurs via a rise in the interest rate on excess reserves (to sterilize the effect of a base-money overhang on credit growth), the ensuing losses for central banks’ balance sheets could be significant.
Eighth, an extended period of negative real interest rates implies a redistribution of income and wealth from creditors and savers toward debtors and borrowers. Of all the forms of adjustment that can lead to deleveraging (growth, savings, orderly debt restructuring, or taxation of wealth), debt monetization (and eventually higher inflation) is the least democratic, and it seriously damages savers and creditors, including pensioners and pension funds.
Ninth, QE and other unconventional monetary policies can have serious unintended consequences. Eventually, excessive inflation may erupt, or credit growth may slow, rather than accelerate, if banks – faced with very low net interest-rate margins – decide that risk relative to reward is insufficient.
Finally, there is a risk of losing sight of any road back to conventional monetary policies. Indeed, some countries are ditching their inflation-targeting regime and moving into uncharted territory, where there may be no anchor for price expectations. The US has moved from QE1 to QE2 and now to QE3, which is potentially unlimited and linked to an unemployment target. Officials are now actively discussing the merit of negative policy rates. And policymakers have moved to a risky credit-easing policy as QE’s effectiveness has waned.
In short, policies are becoming more unconventional, not less, with little clarity about short-term effects, unintended consequences, and long-term impacts. To be sure, QE and other unconventional monetary policies do have important short-term benefits. But if such policies remain in place for too long, their side effects could be severe – and the longer-term costs very high.