Joseph Stiglitz On Rethinking Macroeconomics

♠ Posted by Emmanuel in at 5/11/2011 10:29:00 PM
In the wake of the global financial crisis, various economists have sought means to make macroeconomics and more specifically macroeconomic modelling more faithful simulations of the world economy instead of the other way around. That is, instead of trying to make the world behave more like neat economic models (the Procrustean bed of economic reductionism), several have tried to make macroeconomics resemble true-to-life phenomena more. However, reworking the assumptions of macroeconomics will be an awesomely huge task given how standard economic theory has permeated modern thought. Hence your behavioural economics, neuroeconomics, and all sorts of variations on a theme of making economics a more realistic discipline than assuming hyper-rational actors.

What are we to do with macroeconomics now? Although this theme has been given infinite variations in the wake of the global financial crisis, Stiglitz's spin may be worth considering. Among other things, Stiglitz suggests that we might return to the (good old?) days, presumably when building societies (in the UK) and savings and loan associations (in the US) dominated the home mortgage scene and were ultimately responsible for due diligence as well as risk bearing. That is, instead of passing off risk further down the line to ultimately no-one knows via relentless securitization, consolidating risk assessment and concentration at the traditional lender-borrower interface may be better.

Your mileage may vary; what follows are the abstract and policy implications. Unfortunately, dear readers, the LSE house journal Global Policy from which this excerpt has been taken has now been gated. (The associated PowerPoint slides are still available, though.) So the promotional period ends, but some of the world's leading thinker on global governance--whether you agree with them or not--still populate its pages:

Abstract
The financial crisis made obvious the deficiencies in the standard macroeconomic models, which not only did not predict the crisis, but also said that such events could not occur. While it has long been known that markets are not in general efficient, for example with imperfect information, standard models focused on special cases where the consequent problems did not arise. When market imperfections were introduced, it was done in ways that were ad hoc and/or did not adequately explain deep downturns, such as the Great Recession. This article charts the failures of the standard model, and relates policy failures in the lead-up to the crisis as well as its management to that model’s influence. The standard models focused on providing explanations of normal fluctuations, when what really matters is understanding what causes deep downturns, why shocks to the system get so amplified and why recovery from such events is so slow. (The standard models assumed that the economy was buffeted by exogenous shocks; most crisis shocks are, however, endogenous –‘manmade’.) Four hypotheses are presented about how the structure of the economy changed (sometimes as a result of policy) in ways that increased the likelihood of a large crash with a slow recovery. Finally, the article explains why, in the current context, the policy prescriptions derived from standard models are likely to be misleading.
Policy Implications
  • The focus of monetary policy before the crisis – keeping inflation low and stable – clearly did not suffice to maintain real stability. In the future, monetary authorities need to focus more on the factors that affect the stability of the financial system and credit supply. The deadweight losses associated with the slight misalignment of relative prices associated with low or moderate inflation are miniscule compared to the losses associated with a deep recession.
  • Central banks have at their disposal, in addition to interest rates, a wide range of regulatory instruments. Had they employed these properly, the bubble that caused the current crisis could have been dampened, and the economic consequences of its breaking mitigated. While there may be some costs associated with the use of these instruments, these pale in comparison to the costs of not using them – as the costs of the downturn in the US mount into the trillions.
  • This will necessitate paying more attention to the behavior of the banking system – including tight supervision and regulations, designed to prevent excessive risk taking and excessive interconnectivity, and to encourage banks to focus on lending, especially to small and medium-sized enterprises, which typically do not have access to capital markets. Credit availability may be as important as or more important than interest rates in determining, for instance, investment, especially for SMEs.
  • It is not a surprise that this crisis followed on from financial market liberalization measures taken by the US in recent years; financial crises frequently follow such liberalizations. Globally, financial and capital market liberalizations enabled the ‘made in America’ crisis to spread all over the world.
  • Inherent problems in securitization of home mortgages mean that governments should not count on the restoration of that market – unless it is underpinned with what should be viewed as unacceptable government guarantees. Rather, there should be a return to more traditional mortgage systems (bank based, or the Danish mortgage system).
  • Fiscal policies can be an effective mechanism for reducing unemployment and restoring growth, even in the presence of moderate levels of national debt. Well-designed programs can simultaneously reduce the debt over the long run. By contrast, with interest rates near zero, the contractionary effects of austerity policies cannot be offset by looser monetary policies.
  • Many of the models that became standard in macroeconomics did not incorporate features that allowed them to forecast the downturn (they suggested that such events could not occur), to take actions to prevent such downturns or to respond to the crisis once it occurred. Much of macroeconomics was incoherent – using one set of models, with one set of strong assumptions, to advocate for capital and financial market liberalization, and another set of models to respond to the crises that often follow on from such liberalizations. Models estimated in periods in which firms and households do not face financial constraints and excessive leverage and where central banks are able both to raise and lower interest rates easily do not necessarily provide adequate guidance for behavioral responses in the midst of a deep downturn such as the current one.
  • While it is important to ascertain dynamic responses to current government policies, a wider range of responses needs to be incorporated. Extended periods of unemployment and underinvestment in education and infrastructure can impact future growth and productivity.