A Balanced Debate About Reforming Macroeconomics

IMF Blog, March 22, 2011

Source details

Broad consensus that the macroeconomic models relied upon in the past had failed.
Models failed to predict the crisis; standard models said bubbles couldn’t exist—markets were efficient.
After the bubble broke, models predicted effects would be contained; when effects were not “contained,” models provided limited guidance on how the economy should respond.
Maintaining low and stable inflation did not ensure real economic stability.
Shocks in standard models were treated as exogenous, but the crisis showed shocks can be endogenous.

Fiscal policy can work.
Need to be wary of empirical studies based on circumstances markedly different from the current situation (where households are overleveraged, where interest rates have reached the zero lower bound, etc.).
Central banks will focus on more than just inflation, especially financial stability, but developing an integrated approach will be a real challenge.

The ultimate objective of a central bank is to stabilize the real economy.
Financial and price stability both need to be seen as instruments toward this and other ultimate objectives.
Achieving real stability will require much stronger financial regulation—self-regulation cannot be relied upon because of agency issues and the pervasiveness of externalities.
Real stability will require a full range of tools for capital account management, including cross-border regulations on capital flows.

Consensus that industrial policies have played an important role in enhancing growth.
Other policies, like “rule of law” and macroeconomic stability, are also important.
Discussion went beyond the critique of “picking winners” to analyze externalities associated with learning and development and instances where markets do not work well on their own.

Major failing: micro-foundations in earlier models were not the right micro-foundations for crises.
This crisis was a credit crisis, yet few macroeconomic models modeled credit; banks and securitization were typically not incorporated—even in models used by central banks.
In normal times, credit and money may be highly correlated, but not in crises—the periods when models are most needed for guidance.
There has been extensive modeling of banks and credit creation; task ahead is to incorporate insights from these models into macro-models used by policymakers.

Countries with central banks that were not independent performed so much better than some of those that were—partly because the latter were “cognitively captured” by the financial markets they were supposed to regulate—should perhaps lead to rethinking doctrines concerning central bank independence.
Standard models do not adequately explain the origins of the crisis or the slowness of the recovery.
Despite correction of crisis-related distortions, output has remained substantially below its potential for years, including in the United States, which long prided itself on having flexible labor markets.

Advocates of old policies cautioned against letting the pendulum swing too far: inflation had been a serious problem in the past.
In focusing on other variables, it is important not to lose sight of the risks which high and variable inflation can impose.
Self-regulation failed, but can still be part of an overall regulatory scheme.
Capital flows bring benefits, and these should not be lost sight of.

The conference invigorated a balanced debate about reforming macroeconomics, highlighting failures of standard models, areas of policy consensus, the need for stronger financial regulation, the importance of incorporating credit and banking into macro-models, and the relevance of industrial policy and other growth-supporting measures.

Joseph E. Stiglitz, March 22, 2011