| English
"Куда идет мир? Каково будущее науки? Как "объять необъятное", получая образование - высшее, среднее, начальное? Как преодолеть "пропасть двух культур" - естественнонаучной и гуманитарной? Как создать и вырастить научную школу? Какова структура нашего познания? Как управлять риском? Можно ли с единой точки зрения взглянуть на проблемы математики и экономики, физики и психологии, компьютерных наук и географии, техники и философии?"

«Towards a General Theory of Deep Downturns» 
Joseph E. Stiglitz

Debt policy

The third area of policy is one which has been given short shrift by economists, but seems especially relevant in the current context, where many economies seem plagued with excess leverage. The process of deleveraging and restoring balance sheets is slow. Debt restructuring may be an effective way of restoring aggregate demand more quickly.

Standard economic models would have suggested that such debt restructuring would have little effect: some individuals wealth has increased (those whose debts have been written down), but others have decreased. It is simply a matter of redistribution.

But as we argued earlier, such redistributions do matter. The increase in spending by those whose debt has been reduced (who now face a less binding financial constraint) more than offsets any reduced spending by the creditors. In fact, because of what we referred to earlier as pseudo-wealth, the latter effect may be particularly small: the creditors had, in any case, dim expectations of being repaid. There is a positive increase in the value of aggregate “perceived” wealth.

Inflation used to be an effective way of debt restructuring—in real terms the debts of the debtors got written down. But this approach no longer seems acceptable.

In Stiglitz (2010a) I described a homeowners’ chapter 11 that would have allowed underwater homeowners to do a debt for equity swap. There was enormous deadweight loss from the foreclosure process (and more inequities than even the critics had anticipated, as a result of the robo-signing scandal, where the banks falsely signed thousands of affidavits concerning the debt obligations of homeowners). There was enormous resistance from banks, who were worried about the consequences of the debt write-offs, even if they were simply recognizing losses that had already occurred—accounting “reforms” had allowed the banks to postpone the recognition of these losses.

Contestable democracies

Economists have traditionally divided policy analyses into temporary and permanent changes. But in the real world, there may be a disparity between what the government announces and what agents in the economy believe, and for good reason: in a democracy, each government has only a limited ability to commit future governments, and market participants know this. Sometimes, governments actively participate in the charade: the Bush Administration announced a temporary cut in the estate tax (down to zero in one year). It did this because of budget rules, according to which the country couldn’t afford a permanent cut. But the Bush Administration clearly believed that the political economy was such that once the tax rate reached zero, it would be difficult to raise it again.

In the previous sections, we have considered various measures designed to help the economy get out of a deep downturn. Typically, the measures are announced as temporary. But the effects of the measures depend on beliefs about future policy. Korinek and Stiglitz (2008) have explored how those beliefs affect behavior of market participants, how the knowledge of those behavioral effects the behavior of political actors, and the stochastic equilibrium which thus arises in contested democracies.

III. The capitalist economy as a credit economy

The previous part of this lecture has argued that an understanding of the financial market needs to be at the center of any analysis of deep downturns. Earlier work emphasized the importance of information asymmetries and the role that particular institutions, such as banks, play in dealing with these problems. The failure of much of the standard macro-economic literature to deal with finance and these deep seated information problems accounts for much of the failure of modern macro-economics, and the failure of policy makers to adopt regulatory frameworks that might have prevented the latest deep downturn. For instance, they “bought” into the “completing the markets agenda,” suggesting that structured financed and securitization was leading to a more efficient market—without enquiring whether it in fact was leading to a more efficient distribution of risk and without asking whether there was an attenuation in informational incentives. They seemingly did not understand how new moral hazard problems were being created—for instance, incentives to create defective mortgages were increased, and the market responded to these incentives. Nor did they understand how by opening up new betting opportunities, these “market innovations” were actually increasing market volatility. (Guzman and Stiglitz, 2014, 2015a, 2015b.)

In this part of the lecture, I want to revisit one aspect of the financial sector which has proved particularly problematic: the provision of credit. It was evident after the crisis that private markets did not do a particularly good job in the allocation of credit; and similar criticisms could be raised in earlier crises.

The central role of credit

Traditional monetary economics has focused on money, but in our earlier work, Greenwald and I emphasized what was really important in determining aggregate behavior is credit. Typically, changes in money and credit are highly correlated, which is why econometric analyses typically show a high correlation between money and the level of economic activity. But especially in crises, central banks can expand base money, but the supply of credit may not respond in ways that they normally do, with the result that economic activity does not expand. Keynes was right in emphasizing the inefficacy of monetary policy in such circumstances; but he was wrong in attributing that to a liquidity trap, at least as he conceived of it, due to a horizontal demand curve for money. Rather, we argue that it is related to the unwillingness of banks in such circumstances to increase the supply of credit, even though the constraints that normally limit their supply have been reduced.

Beyond the corn economy: credit creation in a modern economy

In our earlier work, we argued that the traditional argument that the role of banking as intermediating between savers and investors was simply wrong, at least for a modern economy.55 The simple models of financial markets as intermediators provide a description of a corn economy, where some farmers have more seed than they want to plant or consume, and others want to consume/plant more seed than they have. In this simple economy, those with excess seed bring that seed to the banks, which intermediate between the savers and the investors. In this world, a good system of intermediation is simply one with low transactions costs. Markets clear the demand and supply of seed: the interest rate serves to equilibrate the two.56

But this model provides a poor description of our economy today. What enables individuals to spend more than the resources they have available (either for consumption or investment) is access to credit. Credit is different from ordinary commodities. In particular, credit can be created out of thin air. If the bank gives me a piece of paper which others accept, I can buy goods with it, increasing aggregate demand. The bank can create these pieces of paper almost at will. (There are limits, which I shall discuss in a second.) But this is markedly different from the seed economy discussed earlier: the bank then could only lend out seeds if someone else had deposited the seeds—in effect, had lent the seeds to the bank.

Macro-economic consequences of changes in credit availability

With aggregate demand depending on credit availability, changes in credit availability can have macroeconomic consequences. For reasons that we have already explained, adjustments in prices do not instantaneously offset these increases in credit. Moreover, there is no presumption that the market supply of credit will ensure aggregate demand equaling aggregate supply. A key function of monetary policy is to provide the requisite coordination: to ensure that the aggregate demand for goods equals the aggregate supply. In the seed economy, there was no need for this—the interest rate adjusted to make sure that the demand for seed equaled the supply.

Trust as the basis of credit

This poses the central question: what limits the banks’ provision of credit; and what would limit it in the absence of constraints imposed by monetary authorities?

A credit economy is based on trust, and in particular, trust that the money that is borrowed will be repaid; and trust that the money that is received will be honored by others. If a financial institution is trusted, it can create “money” (“credit”) on its own, issuing IOU’s that will be honored by others. (It can thereby increase effective demand; but if the bank is relatively small, it will not take into account these macro-economic externalities of its actions.)

In the “old model” of a credit economy, there was a strong system of accountability for banks issuing IOU’s. There was unlimited liability. If the bank issued loans that are not repaid, the owners of the bank would suffer the consequences.

But the old model often didn’t work. The ability to punish banks and bankers for bad lending was limited. And these problems became worse with the evolution of limited liability, which was necessary and inevitable as the economy became more complex and the banks became larger. It was virtually impossible in a very large partnership for each partner to monitor the others; with unlimited liability partnerships, each partner was fully accountable for the mistakes of all the others. (Greenwald and Stiglitz, 1992.) With limited liability it was difficult—essentially impossible—to hold those in corporations accountable. But the very size of the banks themselves made it also nearly impossible for depositors to effectively monitor the bank. And these problems were exacerbated by increasing complexity of financial system—to the point where not even financial regulators seem capable of assessing the financial position of financial institutions—with multiple instances of a bank’s collapse just shortly after its supervisors had given it a clean bill of health. In short, no one can really monitor the large banks.

The result, even before the financial sector reached anything near the complexity of that of today, was episodic financial panics, as people rush to withdraw funds from financial institutions that had lost their trust. The sudden disappearance of confidence could lead to macroeconomic fluctuations.

In a modern economy, the government stands behind “trust” in the financial system

Today, underlying “trust” in the financial system is the belief that government will come to the rescue, that it is able and willing to do so, and that government is adequately regulating the financial system. Thus, the creation of deposit insurance helped prevent runs, and the strong regulatory system provided confidence in the system as a whole.

After the new regulatory system was created in the aftermath of the Great Depression, we went for decades without a financial crisis. But then two things happened: first, the financial system got better in evading the regulations; and secondly, they exercised their political influence to strip away many of the regulations which had worked so well—arguing in part that because we hadn’t had a crisis, they were no longer needed (when in fact, we had not had a crisis precisely because we had had these regulations). Moreover, they used their political influence to put in place regulators, like Alan Greenspan, the long serving chairman of the Federal Reserve, that didn’t understand the rationale behind regulation, and didn’t believe that regulation was necessary.

Bankers’ moral hazard

The (realistic) belief that government will come to the rescue exacerbates the moral hazard problem—banks know that if they make high interest high risk loans, if things work out well, they walk away with the profits; if things turn out badly, the government picks up the losses. (I have described this as a system of ersatz capitalism—privatizing gains and socializing losses. (Stiglitz (2010a).) I should emphasize that the problem that I am emphasizing is the distortion of banks’ incentives. Some have talked about how deposit insurance attenuates depositor incentives to monitor banks—and have even argued on these grounds that there should be no deposit insurance. I believe this is wrong: no depositor could possibly monitor an institution as complex as Citibank. If the individual attempted to do so, he would have no time to make any money to deposit into the bank. As we noted earlier, not even full time regulators have done a very good job.

More fundamentally, monitoring is a public good—all depositors benefit from ensuring that the bank behaves well; and hence bank regulation and supervision should be publicly provided.57

The moral hazard problem is, of course, worse for financial institutions that are too big, too interconnected, and too correlated to fail. But note that the financial system has incentives to create such institutions. If a bank is too big to fail, it can obtain capital at a lower interest rate (since depositors know that it will be bailed out); and this enables it to expand, unless the government takes countervailing measures, such as charging higher deposit insurance premiums. So too, banks have an incentive to become interconnected, sufficiently enough so that the government cannot allow them to fail.

There are further problems with the too-big-to fail banks: they are too big to be managed, and too big to be held accountable. Thus, even when they were found to have violated the law, enforcement actions were muted, simply because there was a worry about the macro-economic effects of stronger actions. Perhaps most importantly, these big institutions can have disproportionate political influence—shaping regulations, the choice of regulators, bail-outs, and even enforcement actions. Even when the banks were too big to fail, one could have let the bankers, shareholders, and bondholders go. But the big banks were able to use their influence to ensure that this did not happen.

All of this distorts the financial markets: in the absence of tight regulation and offsetting government measures, it leads to excessive risk taking, bad lending, and excessive volatility. Banks expand not on the basis of their efficiency, but simply because of the advantage of size. They undertake contracts (like derivatives) not because they represent efficient risk sharing, but because they help create a system where the bank is too interconnected to fail.

Distortions in the flow of funds

Of course, the belief that a bank will be rescued is tempered by the government’s ability to rescue, giving an advantage to banks from rich countries. The importance of this is suggested by the strong correlation between the CDSs of sovereigns and of the banks within the given country. This helps explain why after the 2008 crisis broke out, money went to the US. It was not because American banks had demonstrated a better ability to manage risk—quite the contrary, they had demonstrated their incompetence. But the US government had already committed itself to a $700 billion rescue, and it was clear that more money would be forthcoming if needed. America had both the resources and the “political economy” (with the financial sector exerting enormous influence, especially on the Central Bank, but also on the Treasury) to ensure a bail-out. Confidence in a rescue was not surprisingly far greater than in the weaker economies of the emerging markets, or even most of those in Europe.

Macro-economic volatility and changes in credit availability

Here, our focus is on credit and macro-economic volatility. Sudden changes in credit availability can result from sudden changes in trust, sudden changes in banks’ perceptions of risk, or sudden changes in banks’ balance sheets (actual and perceived).58 These changes in banks’ balance sheets, in turn, can occur as a result of changes in market prices, in defaults (actual or anticipated), or what Guzman and I refer to as pseudo-wealth: There can be changes in expected future profits as a result of changes in beliefs or changes in opportunities to make profits from others (e.g. in derivative markets) because of differences in beliefs. All of these affect banks’ willingness to lend. In addition, accounting rules combined with banking regulations affect banks’ ability to lend. At various times and for various banks, one constraint or the other may be binding.

The 2008 crisis provided an example of a sudden change in willingness to lend—a liquidity crisis. Banks came to distrust other banks; they knew that they didn’t even know the state of their own balance sheet, so how could they know that of other banks, so the interbank lending market dried up. The crisis also showed the impact of regulatory constraints and accounting rules. The government shifted from mark to market to what I called “marking to hope”—they were allowed to postpone writing down the value of mortgages so long as there was some hope that they might be repaid.

The Fundamental Macro-economic asymmetry

A sudden change in access to credit can give rise to macro-economic fluctuations because of a fundamental asymmetry: The loss of wealth or purchasing power (access to credit) may force those who want to spend more than their income to decrease spending in tandem, while those who gain in wealth (and have access to credit) do not have to increase spending in a corresponding way.

There is a similar situation in the international context which is familiar to all of us; but interestingly the analogy and its implications have not been noted. There has been much worry in recent years about global imbalances. Of course, the sum of the deficits in the world must equal the sum of the surpluses. In standard economic theory, where distribution does not matter and there are no financial constraints, these surpluses and deficits would not matter. So what if some country is lending to another?

The problem (which has been long noted, dating at least back to Keynes) is that there is an adverse effect on global aggregate demand from surpluses. The deficit countries are forced to contract spending, while the surplus countries do not have to expand theirs. A good, well-functioning financial system (working in a world with perfect markets) would easily recycle the surpluses, lending it to others, so that there would be global full employment. But global financial markets often fail to work this way. The surpluses in the surplus countries accumulate in reserves, decreasing global aggregate demand. It used to be that some deficit countries would engage in unbridled spending, and this would offset the deficiency in aggregate demand arising from the surplus countries. But over the past quarter century, deficit countries have become more disciplined; even those that could borrow have curbed their spending. And thus the world has emerged into a situation where there is a deficiency in global aggregate demand. (Greenwald and Stiglitz, 2010a, 2010b.)

Particularly problematic are sudden changes: an increase in the price of oil increases the surpluses of the oil exporters and strengthens the balance sheet of oil exporting firms. It also increases the deficits of the oil importers and weakens the balance sheet of oil importing firms; but the former do not increase their spending as much as the latter are forced to (or choose to) contract theirs.

Inequality gives rise to corresponding imbalances. Those at the bottom are borrowing (running a deficit), while those at the top are lending (running a surplus.) Those at the bottom who see their incomes decline are forced to reduce spending more than those at the top expand theirs; that is, that would be the case unless something else happens, e.g. the country creates a housing bubble. This allows those at the bottom to continue to spend beyond their income—indeed, as we noted earlier in this lecture, the bottom 80% of Americans were spending roughly 110% of their income. This was not, of course, sustainable, and especially so if their spending was based on being able to borrow against their home, the value of which had been inflated by a bubble.

The failure of monetary policy, once again

We suggested earlier, moreover, that the natural adjustment processes may be, at least in the short run, disequilibrating. If the contraction of aggregate demand leads to unemployment and that leads to lower wages (as what, in fact, happened), then aggregate demand is decreased even as the economy attempts to equilibrate.

Easing of monetary policy may not help, or at least help much. Earlier, we explained that what affects borrowers’ behavior is not the T-bill rate (the rate at which the government can borrow), but the rate at which they can borrow, and the spread is an endogenous variable. Moreover, many borrowers face credit constraints, and credit availability is also an endogenous variable. Easing of monetary policy may not help simply because it may not lead to much of an increase in credit or much improvement in the terms at which credit is made available.

Indeed the effect of lowering interest rates T-bill rates may itself be ambiguous. Target savers (those who are saving for purchasing a home, financing a college education, or for retirement) will actually increase their saving and retirees depending on income from government T-bills will reduce their consumption.59 There are, of course, some offsetting effects. Lower interest rates will lead to an increase in the value of shares, and this could be expected to lead to an increase in spending of the wealthy. The question is, how interest-sensitive is consumption of the very wealthy? How much will they increase their spending, especially if interest rate reductions are expected to be temporary? If a new policy regime (e.g. quantitative easing) introduces new macroeconomic uncertainties, it is even possible that the consumption of the wealthy is reduced. As it has turned out, in spite of record low interest rates, the aggregate savings rate has increased, though perhaps not as much as might have been expected or as much as it would have in the absence of the low interest rates. One of the reasons that savings may be lower than one might have expected is that it takes time for individuals to adjust downward their living standards. Some of those in the bottom 80% seem willing to spend beyond their means, so long as banks or other financial institutions are willing to lend to them. If the top 20%, with 40% of US income, save approximately 15% of their income, then that by itself leads to a national savings rate of 6%. If eventually the bottom 80% adjusts to living within their means, with a zero savings rate (i.e. not even paying back accumulated debts), then we can expect the savings rate to increase towards 6%, suggesting that the US recovery will remain weak.

Notice here that what is leading to the weak economy and the limited effect of monetary policy is not the zero lower bound: it is that lowering T-bill interest rates simply doesn’t provide much of a simulative effect to the economy. Banks may not be willing to lend or lend at more favorable terms; and lower interest rates may have ambiguous effects on aggregate consumption.

Alternative approaches to stimulating the economy

If lowering interest rates or lowering wages won’t lead to full employment and the restoration of aggregate demand, what will? For some governments, there is an easy solution: They can create money and credit. Government has the power to print money—and the power to tax, to make good on their promises. It is these powers which lead to trust in the government and in the government’s ability to bailout banks.

In effect, governments have delegated the trust that arises from these powers to the banks, allowing them to profit from the delegated authority. The argument for that is that private banks can do a better job than public institutions, and that there are inevitable “political economy” problems in the public allocation of funds. At best, that would be true if one could align private incentives with public interests, and private returns with social returns. But a central result of the Greenwald-Stiglitz (1986) theorem is that whenever there are information asymmetries and incomplete markets—that is, always, and especially in financial markets—there are pecuniary externalities associated with private actions; and these externalities matter, so that markets are not (constrained) Pareto efficient.

But there are three other systematic market failures associated with the financial system: imperfections of competition, the exploitation of imperfectly informed and often financially unsophisticated consumers, and the special problems of “bail-out” risk that we described earlier.

This delegation has evidently not worked well: transactions costs have been high, there have been enormous market abuses (from market manipulation to predatory lending, from insider trading and connected lending to abusive credit card practices and fraud). The result is that resources have not been well allocated, risks have not been managed well—the markets actually created risks—and the financial sector has played an important role in the increase in inequality. (Stiglitz, 2012a.)

These failures highlight too the political economy problems in the “delegated” solution: for the banks (and especially the big banks) have, as we have noted, used their political power to limit regulation, regulations that might have promoted competition and limited the conflicts of interest, the exploitation of the financially unsophisticated, and the risks imposed on the public. And they have used their political power to extract massive bail-outs from the public. It is evident that political economy problems cannot be avoided. The current arrangements have thus often lead to non-transparent subsidies and distortions.

The standard approach

The standard approach in monetary economics to stabilize the economy focuses on enhancing the ability and willingness of banks to provide credit, through changes in rules that make the constraints on their lending less binding and through open and hidden subsidies that increase their net worth, and thus their capacity and willingness to lend.60 The hope is that they do so, and that the money goes to increase effective demand, rather than purchasing preexisting assets such as land. It is also hoped that they will allocate the funds to uses with the highest social return and that they don’t take advantage of the unwary.

This “solution” hasn’t worked: Banks often haven’t lent (the real source of the “liquidity trap” today), and when they have lent, money hasn’t gone to where it would lead to an increase in effective demand. Indeed, given the risk aversion of banks, it is understandable that they lend against collateral (and some of the rules governing banks may even encourage them to do so). But, lending against collateral means that they disproportionately lend for real estate purchases. Some governments in the past (like that of Thailand) restricted such lending because they wanted funds to go into productive uses, and they didn’t want to create real estate bubbles. They were criticized for doing this—these regulations, it was alleged, were interfering with the market.

We have already explained why these arguments were naïve: they didn’t take into account the multiple market failures that we have noted, the distortions between private and social returns. But leaving it to the market presents a special problem as far as ensuring macro-stability: it is possible that the level of credit necessary to restore the economy to full employment generates excessive increases in asset prices. And indeed, this has often been the case. One was asking too much of a single instrument (the interest rate). Monetary authorities had at their disposal multiple instruments, including those which would limit the extent of real estate speculation. But in the hey-day of “market fundamentalism,” before the 2008 crisis, monetary authorities were discouraged from using these additional instruments. We now know that that was wrong—and the world has paid a high price for this ideologically driven policy. (See Stiglitz 2012b, 2014.)

This analysis helps explain ineffectiveness of monetary policy in the current crisis. Of course, if we had done a better job of fixing the credit channel (a better job at recapitalizing community banks and “fixing” the mortgage market) more of the increased liquidity might have found its way into an increased demand for produced goods, and thus monetary policy might have been more effective.

I should emphasize, the ineffectiveness of monetary policy that I have just described has nothing to do with the traditional Keynesian liquidity trap. Our analysis explains what is really going on: the real constraint is credit availability, and this depends on the behavior of banks. In a deep downturn, it is hard to induce them to lend.61

The current approach not only has been ineffective, but it is also politically unsavory. It has entailed giving money (at below market rates) to those who caused the economic crisis to recapitalize them in the hope that this translates into more lending, and not just into more wealth for the banks. Doing so seems “unjust,” especially since the argument that it was necessary to “save the economy” was never very persuasive, and especially so after it became clear that little of the money got translated into more lending, and some of it went simply to pay huge bonuses to the bankers.62

An alternative solution: Public Lending

There are alternative solutions. One is to induce banks to focus on lending activities. Recent reforms of banking regulations have centered more around preventing the financial sector doing harm to others than ensuring that the financial sector actually performs the social functions which it is supposed to perform. Both incentives and constraints can play a role: Restricting banks’ non-productive activities (like speculation) and providing access to the Fed window on the condition that they expand lending, for instance, to Small and Medium Sized enterprises.

Another is for the government to use its own credit capacity to finance high-return public investments and to address other major social needs, such as those related to growing inequality or climate change.

There already is, of course, some direct lending by the government in many countries, even in more privately oriented systems like the US (for instance, student loans). But this should be greatly expanded, for instance to mortgage lending. The US government already underwrites most mortgages. Most conventional mortgages are written based on standard scoring methodologies, and the relevant information is all within the hands of the government (such as past income and real estate transactions).63 There are large economies of scope both in the processing of this information and in collection through tax authorities.64 The criticism of such lending, as already noted, is that government is not good at lending; but there is an obvious response: neither is the private sector; in fact, arguably no government has ever wasted money at such great cost to society as a whole as the US private financial sector. Government has done a better job at least in some areas and does not seem to have the perverse incentives to behave badly that are pervasive within the private financial sector.

Money rain


There is a final alternative, proposed long ago by Milton Friedman65: Money rain—simply sprinkling money around the economy. It would induce more spending (except under the unlikely condition that prices are adjusted fully, proportionately, and instantaneously). This would not be inflationary, so long as the amounts were appropriately calibrated. (Money rain can be viewed as the limiting case of lending with a minus 100% interest rate.) In many countries (e.g. US) the problem is not an insufficiency of consumption, but of investment, and broad based money rain would restore full employment by encouraging consumption. The solution of government using its credit creating capacity to increase the productive capacity of the economy is more likely to restore the economy to full employment in a way which leads to more sustained economic growth.

IV. The crisis in economics

The 2008 crisis was not only a crisis in the economy, but it was also a crisis for economics—or at least that should have been the case. As we have noted, the standard models didn’t do very well. The criticism is not just that the models did not anticipate or predict the crisis (even shortly before it occurred); they did not contemplate the possibility of a crisis, or at least a crisis of this sort. Because markets were supposed to be efficient, there weren’t supposed to be bubbles. The shocks to the economy were supposed to be exogenous: this one was created by the market itself. Thus, the standard model said the crisis couldn’t or wouldn’t happen; and the standard model had no insights into what generated it.

Not surprisingly, as we again have noted, the standard models provided inadequate guidance on how to respond. Even after the bubble broke, it was argued that diversification of risk meant that the macro-economic consequences would be limited. The standard theory also has had little to say about why the downturn has been so prolonged: Years after the onset of the crisis, large parts of the world are operating well below their potential. In some countries and in some dimension, the downturn is as bad or worse than the Great Depression. Moreover, there is a risk of significant hysteresis effects from protracted unemployment, especially of youth.

The Real Business Cycle and New Keynesian Theories got off to a bad start. They originated out of work undertaken in the 1970s attempting to reconcile the two seemingly distant branches of economics, macro-economics, centering on explaining the major market failure of unemployment, and micro-economics, the center piece of which was the Fundamental Theorems of Welfare Economics, demonstrating the efficiency of markets.66 Real Business Cycle Theory (and its predecessor, New Classical Economics) took one route: using the assumptions of standard micro-economics to construct an analysis of the aggregative behavior of the economy. In doing so, they left Hamlet out of the play: almost by assumption unemployment and other market failures didn’t exist. The timing of their work couldn’t have been worse: for it was just around the same time that economists developed alternative micro-theories, based on asymmetric information, game theory, and behavioral economics, which provided better explanations of a wide range of micro-behavior than did the traditional theory on which the “new macro-economics” was being constructed. At the same time, Sonnenschein (1972) and Mantel (1974) showed that the standard theory provided essentially no structure for macro-economics—essentially any demand or supply function could have been generated by a set of diverse rational consumers. It was the unrealistic assumption of the representative agent that gave theoretical structure to the macro-economic models that were being developed. (As we noted, New Keynesian DSGE models were but a simple variant of these Real Business Cycles, assuming nominal wage and price rigidities—with explanations, we have suggested, that were hardly persuasive.)

There are alternative models to both Real Business Cycles and the New Keynesian DSGE models that provide better insights into the functioning of the macro-economy, and are more consistent with micro-behavior, with new developments of micro-economics, with what has happened in this and other deep downturns. While these new models differ from the older ones in a multitude of ways, at the center of these models is a wide variety of financial market imperfections and a deep analysis of the process of credit creation. These models provide alternative (and I believe better) insights into what kinds of macroeconomic policies would restore the economy to prosperity and maintain macro-stability.

This lecture has attempted to sketch some elements of these alternative approaches. There is a rich research agenda ahead.


2 McConnell and Perez-Quiros (2000) were among the first ones to identify the Great Moderation, which they attributed to changes in inventory management. Indeed, in his presidential address to the AEA, Lucas (2003) went on to describe how the central problem of depression-prevention has been largely resolved. His conclusion was obviously wrong. So too, I suspect were those attributing the Great Moderation to better management by the Central Bank (or even to better inventory management). Inflation was tamed in part by the increasing supply of consumer goods at low prices coming from China. With the decline in the manufacturing sector, inventories had simply become less important. Finally, there was an element of good luck: there was nothing comparable to the oil price shocks of the 70s with which Central Banks had to contend.

3 On March 28th 2007 in testimony before the US Congress, after the bubble had already broken, the

Fed Chairman asserted: “the impact on the broader economy and financial markets of the problems in the

subprime market seems likely to be contained.”

4 See Guzman (2013), who explains the flaws in the results of Aguiar and Gopinath (2006) who had a more optimistic view of the ability of such models to explain the frequency of crises.

5 This kind of hubris was not only exhibited by policymakers, but also by academics. Robert Lucas, in his presidential address to the American Economic Association, famously said:, “[The] central problem of depression prevention has been solved, for all practical purposes, and has in fact been solved for many decades.” (Lucas, 2003)

6 Mendoza and Quadrini (2009).

7 For instance, Popov and Udell (2010).

8 Greenwald and Stiglitz (1987b, 1988a, c,d, 1990, 1993a, 2003), basing their work on models of credit and equity rationing (Stiglitz and Weiss, 1981, 1983; Greenwald, Stiglitz, and Weiss, 1984).

9 The various models assume somewhat different financial frictions, i.e. some focus on borrowing constraints arising from costly state verification (Townsend 1979), others from those associated with adverse selection and moral hazard. Greenwald and Stiglitz in their series of papers also focus on those associated with equity markets. There can be marked differences in results, as we shall comment below. Thus, results based on models with costly state verification may not extend to other forms of financial frictions.

10 In fact, as this paper goes to press, in many European countries, the level of GDP per capita is below the pre-crisis level, and even in the United States and the best performing European countries, GDP remains markedly below the level it would have been had trend growth continued. (See Stiglitz (2015c))

11 These numbers, however, obscure important aggregation problems. The decline in the consumer price index was largely a result of enormous declines in agricultural prices and, to a somewhat lesser extent, housing prices. Manufactured goods prices declined, but to a much lesser extent. See Greenwald and Stiglitz (1987b).

12 As always in complex intertemporal models, things are never quite so simple: if there were variations in future labor supply, it is possible that there could still be changes in lifetime income without changes in the interest rate. That might be the case, for instance, if the government provided a public good which was complementary with private consumption and a substitute for leisure. The standard models, however, do not allow for such subtleties.

13 Even with the same people with the same education, the value of human capital can be lower after the crisis, as the structure of labor demand may change. Differences in skills reinforces the insights of pseudo-wealth theory: agent A thinks the return on activity X is relative large, compared to the return on activity Y; agent B has just the opposite view. The shock may change their valuations in ways which result in a significant decrease in aggregate demand (at any set of wages and prices).

14 The later discussion will identify several other reasons why there may be large changes in aggregate demand.

15 This is partly a result of the fact that bankruptcies sare costly—they are not just redistributive—and partly a matter of accounting and regulation: with mark to market accounting, an increase in bankruptcy probability lowers banks’ net worth, and forces a cutback in lending or an increase in equity. But deep downturns are not a good time to raise equity; existing shareholders are likely to feel that there will be excessive dilution of their ownership claims. They do not take account of the macro-economic externalities associated with the contraction in their lending (just as they do not take account of the macro-economic externalities associated with their excessive lending in the run-up to the crisis). But even in the absence of these accounting and regulatory constraints, the diminution in bank net worth would lead to a contraction in lending, more than offsetting the improvement in the balance sheet of the debtors as a result of the discharge in their debts.