Stiglitz: Old economic models failed us

Needed: a new economic paradigm

By Joseph Stiglitz
August 19, 2010 | Financial Times

The blame game continues over who is responsible for the worst recession since the Great Depression – the financiers who did such a bad job of managing risk or the regulators who failed to stop them. But the economics profession bears more than a little culpability. It provided the models that gave comfort to regulators that markets could be self-regulated; that they were efficient and self-correcting. The efficient markets hypothesis – the notion that market prices fully revealed all the relevant information – ruled the day. Today, not only is our economy in a shambles but so too is the economic paradigm that predominated in the years before the crisis – or at least it should be.

It is hard for non-economists to understand how peculiar the predominant macroeconomic models were. Many assumed demand had to equal supply – and that meant there could be no unemployment. (Right now a lot of people are just enjoying an extra dose of leisure; why they are unhappy is a matter for psychiatry, not economics.) Many used "representative agent models" – all individuals were assumed to be identical, and this meant there could be no meaningful financial markets (who would be lending money to whom?). Information asymmetries, the cornerstone of modern economics, also had no place: they could arise only if individuals suffered from acute schizophrenia, an assumption incompatible with another of the favoured assumptions, full rationality.

Bad models lead to bad policy: central banks, for instance, focused on the small economic inefficiencies arising from inflation, to the exclusion of the far, far greater inefficiencies arising from dysfunctional financial markets and asset price bubbles. After all, their models said that financial markets were always efficient. Remarkably, standard macroeconomic models did not even incorporate adequate analyses of banks. No wonder former Federal Reserve chairman Alan Greenspan, in his famous mea culpa, could express his surprise that banks did not do a better job at risk management. The real surprise was his surprise: even a cursory look at the perverse incentives confronting banks and their managers would have predicted short-sighted behaviour with excessive risk-taking.

The standard models should be graded on their predictive ability – and especially their ability to predict in circumstances that matter. Increasing the accuracy of forecast in normal times (knowing whether the economy will grow at 2.4 per cent or 2.5 per cent) is far less important than knowing the risk of a major recession. In this the models failed miserably, and the predictions of policymakers based on them have, by now, totally undermined their credibility. Policymakers did not see the crisis coming, said its effects were contained after the bubble burst, and thought the consequences would be far more short-lived and less severe than they have been.

Fortunately, while much of the mainstream focused on these flawed models, numerous researchers were engaged in developing alternative approaches. Economic theory had already shown that many of the central conclusions of the standard model were not robust – that is, small changes in assumptions led to large changes in conclusions. Even small information asymmetries, or imperfections in risk markets, meant that markets were not efficient. Celebrated results, such as Adam Smith's invisible hand, did not hold; the invisible hand was invisible because it was not there. Few today would argue that bank managers, in their pursuit of their self-interest, had promoted the well-being of the global economy.

Monetary policy affects the economy through the availability of credit – and the terms on which it is made available, especially to small- and medium-sized enterprises. Understanding this requires us to analyse banks and their interaction with the shadow banking sector. The spread between the Treasury bill rate and lending rates can change markedly. With a few exceptions, most central banks paid little attention to systemic risk and the risks posed by credit interlinkages. Years before the crisis, a few researchers focused on these issues, including the possibility of the bankruptcy cascades that were to play out in such an important way in the crisis. This is an example of the importance of modelling carefully complex interactions among economic agents (households, companies, banks) – interactions that cannot be studied in models in which everyone is assumed to be the same. Even the sacrosanct assumption of rationality has been attacked: there are systemic deviations from rationality and consequences for macroeconomic behaviour that need to be explored.

Changing paradigms is not easy. Too many have invested too much in the wrong models. Like the Ptolemaic attempts to preserve earth-centric views of the universe, there will be heroic efforts to add complexities and refinements to the standard paradigm. The resulting models will be an improvement and policies based on them may do better, but they too are likely to fail. Nothing less than a paradigm shift will do.

But a new paradigm, I believe, is within our grasp: the intellectual building blocks are there and the Institute for New Economic Thinking is providing a framework for bringing the diverse group of scholars striving to create this new paradigm together. What is at stake, of course, is more than just the credibility of the economics profession or that of the policymakers who rely on their ideas: it is the stability and prosperity of our economies.

The writer, recipient of the 2001 Nobel Memorial Prize in economics, is University Professor at Columbia University. He served as chairman of President Bill Clinton's Council of Economic Advisers and as chief economist of the World Bank. He is on the Advisory Board of INET

'Free market' makes health insurance suck

Why Insurers Suck and Five Ways to Make Them Better
By Ezra Klein
February 13, 2008 | Prospect.org

"The state's largest for-profit health insurer is asking California physicians to look for conditions it can use to cancel their new patients' medical coverage," says the first line of this article in The LA Times. What more even needs to be said?

Lets start from a basic proposition: In the current system, insurance companies add negative value, which is to say, they make health care worse, not better. Conservatives often complain that health insurance is not "insurance" in any real sense, it is not protection against unexpected costs, but insulation from largely predictable costs. We know we will need to purchase health care. We contract out with health insurers to smooth those expenditures -- render them predictable and manageable over the course of years, rather than unexpected and crushing in the course of months. That's why we pay insurance premiums so we can one day get chemotherapy, rather than simply paying for chemotherapy.

But "we" here is misleading. Not all of us make this deal with insurers. And among those of us who do make this deal, we make it in different ways, purchasing different levels of insulation, on different time periods. So the insurers, quite naturally, turn their attention to making deals with the most profitable among us, and avoiding deals, or finding ways to break contracts, with the least profitable. They are very innovative in their attempts to do this. But there's nothing good about those attempts. Competition among drug dealers does not aid the neighborhood, and currently, competition among insurers does not aid the ill. Indeed, their inattention to actual care is startling. America, for all its technological advancement, has among the lowest adoption of cost-saving, care-improving, electronic records in the world. That is the fault, in part, of our insurers.

And here's why: It is actually counterproductive for insurers to compete on giving us the best care. It's not simply that they're not doing it, but given the structure of the marketplace, they shouldn't do it. Imagine insurer X creates the best damn diabetes protocols in the country. And they begin advertising this fact. What happens on Day Two? Well, they're flooded with individuals suffering from diabetes, or individuals who fear they will one day be suffering from diabetes. These people, in the current system, are a bad deal. Not only is it near impossible to insure them at a profit, but pooling their costs (which is what insurers do, after all) raises premiums for all the insurer's other customers. When the average customer of an insurer gets sicker, prices go up for all their customers. So the healthy folks contracting with that insurer quit the pool, and go find a cheaper deal, which forces the insurer to raise premiums again, driving out more healthy folks, which forces them to raise premiums again, which drives out more healthy folks, and so on. It's what we call an insurance death spiral, and it ends with the collapse of the insurer.

Given those incentives, insurers cannot compete to offer better care, because if they offered better care, all that would happen is they would attract worse deals. Which is why, in the current system, insurers make things worse. Tyler Cowen has a vision for how insurers could, in a more perfect world, compete to our benefit. And I don't necessarily disagree with it. But let's be clear on what's necessary:

1. Universality: Insurers cannot compete effectively unless everyone is in the pool. If the healthy can leave, insurers cannot compete to offer better care. They'll have to compete to attract the healthiest, which means offering the lowest costs, which means insuring the fewest sick people. The system has to be universal.

2. Community Rating: Insurers cannot be allowed, before offering insurance, to use demographic subslicing to cherrypick the market. That means no more preexisting histories, no complex formulas around age and income and race and region. They offer insurance to anyone who wants it for the exact same price. No exceptions.

3. Risk Adjustment: Merely having everyone in the system won't be enough, and nor will forcing insurers to do away with their most delicate cherrypicking tools. Insurers will just become sophisticated at advertising on G4 Tech TV, and in snowboarding magazines, and in urban centers -- in places, in other words, where the young and the healthy gather. So atop the universal system, atop the community rating, you need risk adjustment, which means either that insurers are reimbursed more for taking on sicker patients, or, my preferred method (and the one used in Germany), insurers with particularly healthy pools pay into a central fund that redistributes to insurers with less healthy pools. At the end of the day, it has to be as profitable for an insurer to insure a sick person as a healthy one.

4. Information Transparency: It needs to be easy for individuals to compare insurers on plan comprehensiveness, price, outcomes, etc. That means we need a marketplace where folks can go to shop for insurers, and they need to have standardized comparisons, or non-partisan rating authorities, providing information they can use.

5. One Market: This is contained in the last point, but there needs to be a singular place, or set of them, where individuals can shop around for insurance. This is hard stuff to find, and harder yet to understand, and real effort needs to go into constructing an easily accessible marketplace that customers can effectively navigate.

There are probably more, but those are the major ones. It's not impossible to imagine a scenario in which insurers actually compete to offer better service, in which the marketplace really does work to the consumer's benefit. That could take a million different forms, from personalized care coordinators to electronic records to online access to your health information to negotiated discounts on gym memberships to a million things I haven't thought of. But none of them happen with any sort of frequency now because insurers operate in a perverse market in which their incentives are to make the system, and our care, worse.