Archive for December, 2010

Keynes and the WW1 reparations

December 27, 2010

This year, Germany finally paid off its old bonds for World War 1 reparations, as Margaret MacMillan has noted in the New York Times.  MacMillan asserts that “John Maynard Keynes, a member of the British delegation in Paris, rightly argued that the Allies should have forgotten about reparations altogether.” Actually, the truth is more complicated.  A fuller understanding of Keynes’s role in the 1919 Paris peace conference after World War 1 may also offer a useful perspective on his contributions to economics.

Keynes became the most famous economist of his time, not for his 1936 General Theory, but for his Economic Consequences of the Peace (1920) and A Revision of the Treaty (1922). These were brilliant polemics against the 1919 peace conference, exposing the folly of imposing on Germany a reparation debt  worth more than 3 times its prewar annual GDP, which was to be repaid over a period of decades.

Germans saw the reparations as unjust extortion, and efforts to accommodate the Allies’ demands undermined the government’s legitimacy, leading to the rise of Nazism and the coming of a second world war. Keynes seemed to foresee the whole disaster. In his 1922 book, he posed the crucial question: “Who believes that the Allies will, over a period of one or two generations, exert adequate force over the German government to extract continuing fruits on a vast scale from forced labor?”

But what Keynes actually recommended in 1922 was that Germany should be asked to pay in reparations about 3% of its prewar GDP annually for 30 years. The 1929 Young Plan offered Germany similar terms and withdrew Allied occupation forces from the German Rhineland, but the Nazis’ rise to national power began after that.

In his 1938 memoirs, Lloyd George tells us that, during World War 1, Germany also had plans to seize valuable assets and property if they won WW1, “but they had not hit on the idea of levying a tribute for 30 to 40 years on the profits and earnings of the Allied peoples.  Mr. Keynes is the sole patentee and promoter of that method of extraction.”

How did Keynes get it so wrong on reparations? In 1871, after the Franco-Prussian War, Germany demanded payments from France, on a less vast scale (only a fraction of France’s annual GDP), while occupying northern France. To hasten the withdrawal of German troops, France made the payments well ahead of the required 3-year schedule, mainly by selling bonds to its own citizens. But the large capital inflow destabilized Germany’s financial system, which then led to a recession in Germany. Before 1914, some argued that such adverse consequences of indemnity payments for a victor’s economy would eliminate incentives for war and assure world peace. In response to such naive arguments, Keynes suggested in 1916 that postwar reparation payments could be extended over decades to avoid macroeconomic shock from large short-term capital flows and imports from Germany.

Nobody had ever tried to extract payments over decades from a defeated nation without occupying it, but that is what the Allies attempted after World War 1, following Keynes’s suggestion. Keynes argued about the payments’ size but not their duration.

Today economists regularly analyze the limits on a sovereign nation’s incentive to pay external debts. In our modern analytical framework, we can argue that the scenario of long-term reparation payments was not a sequential equilibrium. But such analysis uses game-theoretic models that were unknown to Keynes. As a brilliant observer, he certainly recognized the political problems of motivating long-term reparation payments over 30 years or more, but these incentive problems did not fit into the analytical framework that guided him in formulating his policy recommendations. So while condemning the Allies’ demands for Germany to make long-term reparation payments of over 7% of its GDP, Keynes considered long-term payments of 3% of GDP to be economically feasible for Germany, regardless of how politically poisonous such payments might be for its government. Considerations of macroeconomic stability could crowd out strategic incentive analysis for Keynes, given the limits of economic analysis in his time.

Reviewing this history today, we should be impressed both by Keynes’s skill as a critical observer of great policy decisions but also by the severe limits of Keynes’s analytical framework for suggesting better policies. Advances in economic theory have greatly expanded the scope of economic analysis since Keynes’s day and have given us a better framework for policy analysis than what Keynes ever had.


On Keynes and the theory of banking

December 14, 2010

It has been suggested that Keynesian economics remains the best framework that we have for making sense of recessions, but that macroeconomic theory also needs to do a better job of incorporating the realities of finance. There may be a fundamental contradiction between these two suggestions.

In their book Microeconomics of Banking, Xavier Freixas and Jean-Charles Rochet noted that there was no microeconomic theory of banking before the 1970s.  Banks and other financial intermediaries earn their profits by knowing more than depositors about the quality of borrowers’ investments.  So an economic theory of banking requires an ability to analyze transactions among agents who have different information.  Economists first developed such agency theories only around 1970, building on previous advances in game theory.

So John Maynard Keynes’s 1936 General Theory and other classic theories of macroeconomics were developed when there was no real economic theory of banking.  Inevitably this limited the scope of their analysis.  For example, if the 1933 Glass–Steagall Act of banking regulatory reform was essential for halting America’s catastrophic slide into the Great Depression, there would be no way to incorporate that fact into the analysis without an economic theory of banking.

An economic theorist who rereads the General Theory today may be struck by the absence of any serious analysis of how massive bank failures could have been involved in causing the Great Depression.  In chapter 11, Keynes briefly discussed moral hazard in lending, but he had no analytical framework to use these insights, and they tended to get lost in the discussion.

But Keynes was a brilliant observer, even when he could not fit his observations into his theories.  For a contrasting view on the role of banks, look at Keynes’s previous book, his 1930 Treatise on Money.  Near the end of that book, in chapter 37, Keynes made the following observation:

“The relaxation or contraction of credit by the Banking System does not operate merely through a change in the rate charged to borrowers; it also functions through a change in the abundance of credit.  If the supply of credit were distributed in an absolutely free competitive market, these two conditions, quantity and price, would be uniquely correlated with one another and we should not need to consider them separately.  But in practice, the conditions of a free competitive market for bank-loans are imperfectly fulfilled.  There is an habitual system of rationing in the attitude of banks to borrowers — the amount lent to any individual being governed not solely by the security and rate of interest offered, but also by reference to the borrower’s purposes and his standing with the bank as a valuable or influential client.  Thus, there is normally a fringe of unsatisfied borrowers who are not considered to have the first claims on a bank’s favours, but to whom the bank would be quite ready to lend if it were to find itself in a position to lend more.  The existence of this unsatisfied fringe allows the Banking System a means of influencing the rate of investment supplementary to the mere changes in the short-term rate of interest.”

There is an interesting suggestion here that even short-term loans might implicitly depend on long-term relationships between investors and financial intermediaries.  Such an idea could be the basis for a theory of macroeconomic fluctuations in which bank failures could affect investment.

In 1936, however, Keynes could not build a theory in which monetary policy could affect aggregate investment other than through its effect on the interest rate.  His 1930 observation got lost in his subsequent analysis because it did not fit into his analytical framework.  He had no way to answer the obvious question: If so many eager qualified borrowers are unable to get loans at the current interest rate, why don’t banks offer to lend to them at a higher interest rate?  Today economists understand how such credit rationing can be derived from considerations of adverse selection or moral hazard in borrowing.  The classic introduction to the subject is by Joseph Stiglitz and Andrew Weiss in 1981.

Understanding the language of force

December 8, 2010

Sometimes it seems that they only understand the language of force. We have learned from history that the appeasement of an aggressive adversary can be a disastrous mistake, with our concessions only encouraging further attacks on our communities. To deter them from attacking us, we need armed strength, and our leaders must demonstrate the resolve to use it when necessary.” What should we say to someone who describes the Israeli-Palestinian conflict in these terms?

A game theorist is trained to look at conflict problems from both sides, assuming that people on both sides are rational and intelligent. I have tried to write the above quote as one that many Israelis and Palestinians might consider a fair description of their situation, symmetrically identifying themselves as “we” and the other side as “they”, but the symmetry of this view is probably not common knowledge. In particular, many may not understand the other side’s fear of appeasing an aggressive adversary. Such misunderstanding can undermine hopes for peace.

In the above quote, the response to our armed strength that “we” seek from “them” is, in a word, appeasement. We want them to appease us. But why should they not fear that concessions to us could encourage our greater ambitions, inviting further invasion of their communities? And if the demand for armed vigilance on each side is matched by a fear of appeasement on the other side, how can the two sides ever escape from the long war of attrition?

We must think more carefully about the logic of deterrent strategies. Our strategy to deter potential adversaries must have two parts: a threat that we will fight them if they attack us, and a promise that we will be good restrained neighbors if they accommodate us. The difference between our threat and our promise is what encourages them toward accommodation. For our deterrent strategy to be effective, our potential adversaries must understand and believe both the threat and the promise.

Failing to credibly communicate the threat is naive appeasement. Our potential adversaries must not think that we are the weak type of people, who lack resolve to respond forcefully against aggression. To prove that we are not weak may require costly signals of our resolve, many of which have become too familiar: sending out young men on deadly missions against the other side.

But deterrence can fail also if we do not credibly communicate a promise that differs from the threat. If they believe that we are an aggressive type, who cannot restrain ourselves from invading their communities further at any feasible opportunity, then they will feel driven to seek militant leaders against us, and then we will be locked in conflict with them.

How can we demonstrate to them that we are not such an aggressive type? This is a very serious question, because everyone knows that aggressors may try to mask their intentions with honeyed words of peace. The point is not to convince ourselves of our own moral purity; the goal is to convince our adversaries that they can safely make peace with us.

We can effectively signal our restraint by articulating clear strategic limits that verifiably constrain our actions in the conflict, and by showing real understanding and respect for justice as our adversaries see it. Credibly communicating our promise of restraint to a suspicious adversary can be a long and difficult process, but it is an essential part of effective communication in the language of force.

That is the theory. Today we learned that Israel has resisted intense American pressure to freeze expansions of its settlements in the West Bank. It is hard to see this decision as a signal of restraint. Indeed, it seems just the opposite. In rejecting its strongest ally’s interpretation of the legal limits on its expansion, Israel seems to have given a costly signal of an inability to restrain expansionist forces in its political system. Nobody can enter into a treaty without confidence that the other side will accept a mutually agreed interpretation of its limits under the treaty. As a costly signal that reduces the other side’s willingness to make peace, this decision may be less stark than missiles from Gaza, but it is only a matter of degree.