Conflicting Thought on Eradicating Deflation

Fellow, RIETI

During the protracted recession that gripped Japan after the collapse of the bubble economy, there have been various debates over economic policy. In the end, the point of conflict of these arguments can be summed up as: "which should come first, propping up the economy or implementing reforms?" A recent example would be the debate over how to deal with deflation, which pits those who argue that "deflation should first be eradicated by having the Bank of Japan implement more aggressive monetary easing" against those who maintain that "deflation will not abate without financial system reform through nonperforming loan disposal and corporate reorganization."
This debate has often descended into emotional slander, but I have long wondered why this should happen. It seemed different from simply being the result of a heated debate turning emotional. I began to think that this antagonism in fact arose from some confrontation between root values or philosophy.
This conflict of values is easy to understand when considered using the framework of "liberalism" vs "constructivism." This is a framework put forward by F. A. Hayek, winner of the 1974 Nobel Prize in economics (The Constitution of Liberty, Law, Legislation and Liberty).
In this column, I would first like to sort out the confrontation of values behind recent economic debates (especially those regarding deflation) in line with Hayek's framework of liberalism and constructivism.
Then, I will investigate why such a conflict of values becomes tangible when discussing monetary policy issues and point to the lessons to be learned for future policy philosophy.

1. Spontaneous market rules cannot be constructed by reason

Liberalism and constructivism

Liberalism is the belief that competition among free individuals is the greatest force that shapes modern social order. It also places the greatest value on securing free competition. The important conclusion drawn from this line of thinking is that the rules and regulations for competition must be observed. This is because free competition without rules would lead to complete chaos; a Darwinian jungle where injustices can pass unmarked. One of the rules of economic competition is that companies and banks that have been defeated in competition must leave (or be revived through necessary structural adjustments). Those who place priority on structural reform such as nonperforming loan disposal ultimately want this rule to be observed. In this sense, proponents of structural reform can be called liberals.

In contrast, constructivism believes that order in modern society can be artificially constructed through human reason. This is the belief that a desirable social order can be achieved by having neutral social planners with a high level of reason artificially distribute the resources of society as a whole. It is, so to speak, the belief that constructors (the government) can control society just as scientists can control things in nature. The recent argument for eradicating deflation is that "the Bank of Japan can control the inflation expectation of the public if it implements appropriate measures. Therefore, deflation can be dispelled through monetary policy." This argument, as I will mention later, has a very high affinity with constructivism.

I will leave the deflation debate for the next section, and here I would like to study the fundamental idea of liberalism - spontaneous market rules - and examine its essential difference from constructivism. Hayek, who was an advocate of liberalism, concludes that the idea of constructivism itself "is a superstition in the strict sense of the word." Let us look at the grounds for this argument.

Signs of the confrontation between liberalism and constructivism are said to have already been visible in ancient Greece. It has also repeatedly appeared in modern civilized society as a point of argument regarding thought and policy. According to Hayek, the unavoidable result of pushing forward with constructivism is totalitarianism (fascism or communism), and it may be said that World War II and the Cold War were the biggest conflicts between liberalism and constructivism. In the field of economics, it may be said that neo-classicalists tend to be liberals, while Keynesians rather have a tendency to be somewhat constructivist. The difference between liberalism and constructivism is the difference between whether social order is viewed as "order of the market" (a spontaneous structure brought about by market competition) or "order of the organization" (an organizational structure created through artificial construction).

Market order was not something that someone constructed, but the result of unpredictable competition. Individual market participants (individuals, companies, banks, etc) have no way of possessing all the information regarding the market, and act to achieve their goals based only on the information around them. The reason why the separate actions of market participants together form one fixed order is because it is (quite) certainly guaranteed that "all market participants are following the rules of the market." Some examples of the rules of the market are, "how to handle private property is up to the owner, and others cannot do as they please with it," or "those who violate others' property (those who cannot repay debts) will be punished in such and such a way."

By trusting the fact that "the rules are being observed," individual market participants can dispel future uncertainty and predict the results of their actions with a fair degree of certainty. The role of market rules is to provide certainty to the future "expectations" of individual market participants and create an environment where they can do their best. As a result of their doing their best, a spontaneous market order can be created.

Where do the rules of the market come from?

So then, who made the rules of the market in the first place? We would like to assume that "someone" (the government, the Walrasian auctioneer, etc) created the market, but Hayek's answer is "nobody." The rules themselves are selected through market competition, and the most widely accepted rules survive. This mechanism is similar to that of the evolution process of living organisms. Not only the structure of order in the market (such as the distribution of companies and industries) but also the rules of the market themselves are formed as a result of competition. Competition takes place in line with rules, but market participants on occasion accidentally alter part of those rules in line with the circumstances they find themselves in. Under this mechanism, the changes adopted by the market participants that succeeded as a result of competition become widely accepted and the rules of the market evolve.

But if the rules of the market are spontaneous and are not forced upon us by someone, why do individual market participants follow them in the first place? Hayek stresses that this is "not because the rules are rational." Rational here means being able to be deductively constructed through human reason, excluding custom and tradition. For example, just as the Japanese and American economic systems can coexist, there is no one and only model for the spontaneous rules of the market. Why it has to be this rule (or set of rules) cannot be deduced by the reason of individual market participants alone, and if I were to be forced to explain, I would say that it is only because it would be more profitable to follow those rules because everyone else is following them. In other words, the rules are not deduced from human reason, but they come to be followed because people think rationally based on the assumption that other people are observing them.1

That is, the rules of the market are the traditions and customs of economic activity themselves. In Hayek's line of thinking, free market competition becomes possible when individual market participants respect and observe traditions and customs (things which at first glance may seem irrational) that cannot be deduced from reason. Individual market participants use their reason amid tradition and customs, and it is not that reason functions by transcending such existing value systems.

"What [constructivist] rationalism ... regard[s] as senseless and meaningless formations due to accident or human caprice [in other words, market rules as tradition and custom] turn out in many instances to be the foundations on which our capacity for rational thought rests." (Law, Legislation and Liberty, square brackets are mine.)

A point of greater importance is that "it is impossible to fully state in words all the market rules that people are following." This is because current rules are temporary, and they evolve in line with new situations. This is the same as, for example, the way in which the Civil Code and the Penal Code get new interpretations through judicial precedents or have articles revised in order to deal with new incidents. Or we can say that the evolution of market rules is similar to progress in the natural sciences, such as physics. Current physical theories are temporary, and if new experimental facts are discovered, theories will be modified to explain them, or will be replaced by new theories. No one knows the "ultimate truth" of physics, but yet we use this "provisional" physics to get by in our daily lives. Hayek argues that the same can be said for market rules.

It is very difficult to artificially control market rules that are spontaneous, temporary and change accidentally and the market order created through those rules. What happens when we look at constructivism based on this view of market order? The major problem with constructivism is its conviction that "spontaneous market order can be substituted by an organizational order that has been artificially constructed through reason."

Human instinct repels market competition

Hayek says the model for organizational order lies in the tribal societies of primitive hunting peoples (or the village communities of agricultural peoples). In small communities, all members know each other and their leader distributes resources and work among them. This organizational order has been passed down in an unbroken line to modern times as the organizational principles that guide the military, government ministries and private corporations. There is no problem with the idea that organizations like companies should be constructed rationally through reason. However, there is a big gap in saying that the same holds true for market order.

When market rules develop spontaneously and are not deduced from reason, and when no one knows the ultimate appearance of market rules and everyone accepts them as a temporary truth, it is theoretically impossible to construct such rules artificially. Therefore, (because market rules cannot be artificially constructed) attempts to revise market order using artificial designs in the end leads to the distribution of resources based on individual, specific orders from the constructor. Because in fact this constructor only has limited knowledge, his or her distribution of resources has an unpredictable impact on market order and results in a distribution that many feel is "arbitrary and unfair." Economic efficiency will also decline in comparison with market order.

This is Hayek's criticism of constructivism (the idea that constructing an artificial social order can achieve a better society than market economics).

Finally, Hayek offers an interesting diagnosis of why the conflict between liberalism and constructivism never goes away. It runs counter to primitive human instinct to observe spontaneous market rules that have not been forced upon us and enter the world of free competition. There is no mistake that modern industrial society has developed through the expansion of division of labor through free competition, but in a world of market competition individual people have a greater sense of alienation and feel more stress by being forced to accept self-responsibility. The activities of the individual are linked to those of an innumerable number of people through market transactions (this is the driving force behind the development of civil society), but in the end these are only faceless business relationships, and the person cannot feel any warm, human bonds. Amid such a situation, the desire for the sense of unity of a primitive community and the security of obeying an authority increases, and this in turn breeds repulsion from market order and strong support for a constructivist organizational order.

This is a phenomenon that repeatedly occurs whenever the economic situation deteriorates. It is also an important problem that must always be kept in mind when discussing economic policy.

2. Nonperforming loan disposal and liberalism

What the Japanese economy learned over the past decade

How can the problem of nonperforming loan disposal be sorted out from the above view of market economics?
It may be said that the methods employed by authorities in the first half of the 1990s, when the nonperforming loan issue started to become serious, were indeed constructivistic. Under the leadership of financial authorities, nonperforming loans were to be disposed over time, gradually and with premeditation. Based on the principles of the Commercial Code, nonperforming loans must be promptly written off and disposed as they crop up, and so this policy was the equivalent of temporarily suspending application of the Commercial Code (market rules) to the nonperforming loan issue and proceeding with loss disposal (resource distribution) on a case-by-case basis under the management of the government.

It may seem that not promptly disposing of sour loans based on the rules and instead resolving the matter over time based on a plan constructed by the government was a rational decision considering the possible impact on banks and the economy. It may seem especially correct given that it was a decision made in the first half of the 1990s, when a protracted economic downturn had not been foreseen.

However, from the abovementioned view of market order, this policy decision should be viewed as having excessively taken light of the impact of creating a huge exception to the rules of spontaneous market order, or, an overestimation of the constructing abilities of the government.

Even if we disregard the causal relationship of whether the delay in nonperforming loan disposal brought about the prolonged recession, it is believed that the failure to swiftly dispose of sour loans in line with rules greatly damaged the reliability of the market rules of the Japanese economy as a whole. In other words, because the rule dictating "what happens when debts cannot be repaid" was not followed, the extreme uncertainty that "even if debts cannot be repaid, the borrower may survive or be forced to go bankrupt, based on an arbitrary decision by the government or the banks" was born.

If the role of market rules is, as Hayek says, the removal of future uncertainties faced by individual market participants and the increasing of the certitude of expectations, the creation of a huge exception in market rules regarding nonperforming loan disposal must have increased the uncertainty in all sorts of economic activity and thrown a wet blanket on economic transactions. For example, when a company plans to undertake a new business, it will be haunted by various suspicions that were unthinkable when market rules were trusted, such as "banks might protect the other party in the transaction if the business fails, and we might suffer damages."

The seriousness of the adverse effects that the destruction of trust in rules has on market order cannot be calculated in advance through reason. This is indeed the core of Hayek's warning regarding constructivism, and what it took the Japanese economy 10 years to learn.

At present, there is probably little objection to the need to rapidly dispose of nonperforming loans. However, the reason to do so is not just the economic reason that any further delay will cause the deflationary recession to drag on.2

There is a need to restore trust in market rules, and to do so, we must regain the attitude of respecting and observing market rules (which may at first glance seem as though we can dump them if we think rationally). This, rather, is the significance of proceeding with nonperforming loan disposal.

From such a viewpoint, nonperforming loan disposal should proceed in a way that clarifies the rule of "what happens to companies and banks that fail in their businesses," and in other words, the main part of the work should be finding an answer to the issue of responsibility. How should the issue of the individual responsibility of borrower firms, lender banks and financial administrators be settled?3 Clarifying this should be the first step in restoring trust in the warped and confused rules of the market. With the economy showing strong indications of recovery and fears over a financial crisis having receded, now is indeed a good opportunity to review and calmly summarize the details of past nonperforming loan disposal efforts.

Confusion in thought over capital injections into banks

Two counterarguments can be expected to the argument that nonperforming loan disposal is necessary to restore trust in market rules.
The first is that the market rule that "nonperforming loans should be promptly disposed of" is a Western one, and that there should be a different way for Japan to address the problem. However, this is not correct. The market rule that was destroyed by postponing nonperforming loan disposal, namely, the common understanding that "this is what happens when debts are not repaid," is a commercial ethic that has developed spontaneously in Japan since the Meiji Period or the Edo Period, or perhaps even before then. It is just that because the banking system and the accounting system are Western in origin, the mistaken image that this commercial ethic is also a Western rule is prevalent. The rule that was broken when nonperforming loan disposal was delayed was a commercial ethic that has existed in Japan since olden times, and it is this commercial ethic that we need to restore. The need to dispose of nonperforming loans is completely unrelated to the debate over whether Western rules should be introduced to the Japanese economy.

The second counterargument may be that "if banks become undercapitalized as a result of rapid nonperforming loan disposal, it will become necessary to inject new capital using public funds. Wouldn't this be state intervention and a constructivist policy?" However, capital infusion is the natural consequence of a policy decision to not have depositors shoulder the burden of losses when a bank loses their property, and is unrelated to constructivism or state intervention. When a bank damages depositors' property due to nonperforming loans, it is theoretically possible to dispose of the bank in a way in which depositors directly shoulder the losses. However, it may be said that having the government protect depositors to save the payment and settlement system is also a rule that has come to be accepted to a certain extent today. As a result, capital injections using public funds means that taxpayers take on the losses that depositors would have suffered, and this is consistent with the rule that banks should take responsibility when they damage depositors' property. Whether a bank should be liquidated or resuscitated after depositors' losses are covered with public funds is a separate policy decision.

The reason why the government's current capital infusion policy is difficult to understand is because it is based on the ambiguous idea that capital will be injected into "healthy banks." The phrase "healthy banks" gives the impression that they are not insolvent, but if they were not insolvent, they would not need capital injections. Authorities should clarify their policy idea by saying that capital will be injected into "banks that are currently insolvent but are expected to regain high profitability in the future."

Can it not be said that so far, the speed of financial revival has been slow because the ambiguity and confusion of policy ideas regarding capital injection has led to a lack of public support for the scheme, resulting in a delay in necessary capital injections? I believe it is necessary to sort out the policy ideas regarding capital injection in preparation for any future financial crisis.

3. Controlling inflation expectation - the trap of constructivism

The two factors that decide price levels

In terms of nonperforming loan disposal, we have been able to say that from a liberal point of view, Japan must strive to revive market rules. However, in dealing with deflation, the Bank of Japan's monetary policy is shouldered with a difficult problem.

First, let me present my basic views regarding the Bank of Japan's monetary policy. I believe that while quantitative easing must continue in order to support the economy, this alone cannot fully control the economy. My basic position is that to secure sustainable economic recovery, nonperforming loan disposal must be swiftly completed while quantitative easing continues. In this section I would like to study the recent theory that it will suffice for the Bank of Japan to pursue an aggressive monetary easing policy to eradicate deflation (secure economic recovery).

A situation seen in present-day Japan, where deflation continues for years amid nominal short-term interest rates of zero percent, is abnormal and has never before been assumed by economics. What sort of role the monetary policies of the central bank can play in such a situation is an important theme for academic research. It is natural that economists around the world have begun studying the way monetary policy should be implemented under zero interest rates or deflationary conditions, and their serious research efforts deserve respect.

However, how has it come to pass that those who are not experts in monetary policy are passionately advocating that "deflation can be eradicated if the Bank of Japan implements an aggressive monetary policy" and that such arguments are gaining broad support among nonexperts? I suspect that many of these nonexperts are being gripped by constructivist passion.

Let us look more closely at the argument that Japan can shake off deflation through aggressive monetary policy. The reason the issue of deflation is difficult is because price levels are decided by a combination of two factors - base money, which is the product of artificial construction (and which the Bank of Japan can control) and the velocity of money, which is determined by market order. The relationship among prices, base money and the velocity of money can be shown as follows:


Here P stands for price levels, Y for real gross domestic product (the amount of goods), M for base money supplied by the Bank of Japan and V for the velocity of money.

The amount of base money can be more or less decided by the Bank of Japan. However, the Bank of Japan cannot determine how rapidly the base money supplied to the Japanese economy is distributed throughout the entire economy. What decides this speed is the total of transactions among banks, companies and consumers, and indeed it may be said that market order decides the velocity of money. In order to simplify discussions, the theoretical model used in economics usually assumes that velocity V is constant. In other words, when considering the effects of the base money M that the Bank of Japan controls, it is a theoretical prerequisite that the velocity of money V, which is decided by market order, does not change.

Can the Bank of Japan control "inflation expectation"?

However, in the actual Japanese economy, V has been declining every year throughout the recession since the burst of the economic bubble, and especially since the late 1990s, when the Bank of Japan sharply increased base money supply, V has dropped rapidly, as if to cancel out the effects of monetary easing policies.

Diagram: The velocity of money since the 1990s

The popular argument for eradicating deflation is that (when production Y and velocity V are fixed) prices P should rise when base money M grows, and so if the Bank of Japan boosts M through monetary easing, the public should have inflation expectation. If inflation expectation rises (under the special circumstances of nominal short-term interest rates standing at zero percent), such expectation becomes real inflation.

But what would happen if velocity V falls because the public harbors the pessimistic expectation that "deflation will continue"? In such a case, there is the possibility that prices P will not rise even if the Bank of Japan increases base money M, so the deflation expectation of the public cannot be dispelled. In other words, if velocity V, which is determined by market order, changes, the policies of the Bank of Japan cannot always control the public's "inflation expectation." 4

As can be seen from the above argument, the problem with the theory for eradicating deflation is that it assumes that there will be no decline in the velocity of money V when the Bank of Japan implements its policies, or, even if there is, it will be negligible. In other words, it underestimates the changes in market order.
The assumption that "velocity V is fixed" is one used by theoretical researchers for the sake of convenience when formularizing their theories. As seen from the above, it does not hold water in reality. What can it mean when this assumption made for convenience's sake continues to be tacitly assumed even in discussions concerning policy proposals?

I believe this is the very same kind of jump in logic made by constructivism. Constructivism believes that "organizations such as corporations can and should be controlled by rational construction. Therefore, order in the economy as a whole, including market order, should also be controllable through a constructed plan." The argument that deflation can be eradicated also believes that "base money M can be fully controlled by the Bank of Japan through rational construction. Therefore, price levels P should also be controllable in line with the Bank of Japan's constructions." However, P is affected by velocity of money V, which is determined by market order. It cannot be helped for the aforementioned argument for deflation eradication to be called a jump in logic so long as it is not put forward after clarifying the movements of V (in other words, market order).

Shaken trust in market rules

Studies in economics have yet to sufficiently explain why the velocity of money V continues to fall in present-day Japan. Perhaps it is declining in line with some law that renders the monetary easing of the Bank of Japan ineffective. When considering such possibilities, it cannot be declared from the position of expert researchers at this stage that "deflation can be eradicated solely through aggressive monetary easing."

The problem is that the argument that "deflation can only be resolved through monetary policy" has gained strong support among nonexperts with this jump in logic overlooked. It is also interesting to note that both conservative and progressive debaters alike support this argument. I believe this shows that amid the sense of despair over the economy, the illusion of constructivism - the desire to flee from market rules and place oneself in a primitive and methodical organizational order - has spread among the ranks of Japan's critics.

We should also note that nonexperts who support the argument for eradicating deflation are generally skeptical or critical of the idea of following market rules on such matters as nonperforming loan disposal. The characteristic of constructivism is the belief that spontaneous market rules can be taken lightly and that everything can be replaced by control through rational constructors. From the constructionist viewpoint, it is only natural to think that there is no rational reason to simply follow customary market rules (such as disposing of nonperforming loans in line with rules) and that it is ridiculous to do so. This is where the snide argument that "there is a cleverer way than just honestly following rules" comes from.

However, their argument closes its eyes to the fact that neither the government nor the Bank of Japan can control the velocity of money (market order) and is based on a jump in logic - indeed, can we not say that they have been captivated by constructivist passion?

It may rather be the case that restoring trust in market rules is the shortcut to eradicating deflation. Let us suppose that trust in market rules was destroyed due to such factors as the delay in nonperforming loan disposal, and that as a result, economic activity withered and the velocity of money V is falling. If this is true, disposing of nonperforming loans in line with rules and restoring confidence in the rules may lead to a rise in the velocity of money V. In such a case, prices P will rise along with V. Japan may be able to shrug off deflation through a rise in V, without having to sharply increase base money M.

4. Lessons for the future

The above argument deconstructed the recent debate surrounding deflation and reforms (especially nonperforming loan disposal) from the viewpoint of a conflict in thought that has existed since ancient times - liberalism versus constructivism. So long as market rules that bring about economic development are uncomfortable to human instinct while the concept of "constructed organizational order" remains attractive, this conflict will likely continue. What we need to be careful of is not to make important policy decisions while being captivated by constructivist passion and overlooking leaps in logic.

In the near future, the same sort of debate will likely be repeated when the crisis of fiscal collapse (a government bond crash) becomes a matter of great urgency. Amid the present situation in which the fiscal deficit and outstanding government bond issues continue to surge, there are only two options available. One is fiscal reconsolidation through spending cuts and tax hikes. The other is to adopt inflationary policies through the unlimited purchase of government bonds by the Bank of Japan.

In the sense that the government itself will fulfill its obligations as a market participant in line with rules, fiscal reconsolidation can be said to be a liberal policy that respects rules. On the other hand, trying to resolve the fiscal problem through the Bank of Japan's purchases of government bonds and ensuing high inflation can be seen as a constructivist approach that aims to overcome the problem by only controlling macroeconomic variables.

Incidentally, one famous example of an attempt to fix fiscal problems through inflationary policies is that of the Latin American countries in the 1970s and 1980s. Inflation progressed autonomously once it was unleashed, and finally these countries were tormented by high and protracted spontaneous inflation that could not be explained by fiscal problems alone. In other words, these policies resulted in the emergence of a change in market order (the firm establishment of autonomous high inflation) that policy constructors could not foresee at the outset.

The policy debate over whether to choose fiscal reconsolidation or inflation will likely take on aspects similar to that of the current argument surrounding deflation. What will then be needed is to know the limits of constructed policies and to make cautious policy decisions after humbly accepting the fact that we cannot change market order in whatever way we like through policies.

>> Original text in Japanese

* Reprinted from Nihon Keizai no Ronten (Diamond Inc.)

June 2004 Nihon Keizai no Ronten

  1. From game theory-type observations on the actions of such market participants, Professor Masahiko Aoki of Stanford University (comparative institutional analysis) stresses that various market rules (systems) develop in different areas.
  2. There has been recent progress in both theoretical and positive research regarding the link between nonperforming loans and the protracted recession, and it seems very certain that nonperforming loans are the cause of the economic downturn. However, it has not been 100% proved. There are still many who would object to the idea if this causal relationship is presented as the only reason why nonperforming loan disposal should be pursued.
  3. What is important is to ensure that the settlement of responsibility issues does not become arbitrary. Therefore, there is no need for punishments to be severe, and exemption of responsibility based on rules may also be acceptable.
  4. Experts may say that "because there is a lower limit to velocity of money V, equilibrium with deflation expectation will not satisfy transversality conditions if the Bank of Japan continues to increase base money supply for a long period of time. Therefore, at the point of equilibrium the general public will no doubt hold the expectation that 'inflation will occur'." However, the prerequisite for this logic to hold water is if the public has the "expectation" that "the Bank of Japan will maintain its easy money policy for a long time." If the public expects that "the Bank of Japan will no doubt move to tighten monetary policy in the future," it would not be strange for deflation expectation to continue. In other words, the "expectation" that the Bank of Japan's easy money policy will continue for a long time must first exist in order for the public to have inflation "expectation." So far, no one has been able to propose a clear-cut and reliable method by which such expectation toward monetary easing policies can be fostered (with the exception of methods that combine foreign exchange policy and fiscal policy).

August 4, 2004

Article(s) by this author