Skip to Content

Fourth Session : The Expert Group on the Recent Trends in International Finance

Councilon Customs, Tariff, Foreign Exchange and Other Transactions
Subcouncilon Foreign Exchange and Other Transactions

Fourth Session : The Expert Group onthe Recent Trends in International Finance

Date: Wednesday, 16 April 2003 10:00-12:00
Venue: Ministry of Finance (ConferenceHall)


Opening Remarks by Chairman of the Expert Group, Professor Naoyuki Yoshino, KeioUniversity
(1) "Deflation, Globalization and The NewParadigm of Monetary Economics"
Presentation by Professor Joseph E. Stiglitz,Columbia University
Comments by Mr. Haruhiko Kuroda, Special Advisorto the Cabinet
Comments by Professor Naoyuki Yoshino, KeioUniversity
(2) Question-and-Answer Session and Discussion
3.Closing Remarks by Professor NaoyukiYoshino

Mr. Naoyuki Yoshino, Chairman of theExpert Group on the Recent Trends in International Finance:
It is almost 10:00 so we would like to now hold the fourth session of the ExpertGroup on the Recent Trends in International Finance of the Subcouncil on ForeignExchange and Other Transactions of the Council on Customs, Tariff, ForeignExchange and Other Transactions. Seated next to me is Professor Joseph Stiglitz,who is the 2001 Nobel Laureate of Economics. He is currently a professor atColumbia University and we would like to first invite Professor Stiglitz to makea presentation.

Since this is indeed a valuable opportunity, we have also invited participantsfrom outside the expert group and I would like to apologize for the fact thatsome of you might be rather squeezed. As indicated on the panel, ProfessorStiglitz's presentation title is "Deflation, Globalization and the NewParadigm of Monetary Economics" and he will speak for roughly 16 minutes.After that, Mr. Kuroda, who is former Vice Minister of International Affairs forthe Ministry of Finance (MOF) and currently Special Advisor to the Cabinet willspeak and then I will provide some comments. Finally, we would like to entertainquestions and comments from the floor.

I am sure that you know who Professor Stiglitz is, but I would like to brieflyintroduce his CV to you. After graduating from Amherst University, he went toMIT Graduate School and then he studied in Cambridge University to obtain hisDoctorate in the United Kingdom. In 1967, he first started Yale University, andin 1970 he became a full professor at Yale. Subsequently, he also serves asProfessor at Stanford, Oxford, and Princeton Universities. In 1979, he wasawarded John Bates Clark Prize for being the young economist who has made thegreat contribution to US economics. Students of economics have all had theopportunities to read the papers and thesis by Professor Stiglitz, and invarious areas he has actively published the papers. Also within the USgovernment, in March 1993 he participated in this Council of Economic Advisors (CEA)of the Clinton administration and from June 1995 he became the chairman of theCEA and he has been closely involved in the management of economic publishing inthe United States. In February 1997, he retired as CEA chairman and becameSenior Vice President and Chief Economist for the World Bank, a post he had helduntil January of 2000. In the year 2001, he was awarded the Nobel Prize forEconomics by contributing his information in Economics and currently he is aProfessor at Columbia University. So without further ado, I would like to inviteProfessor Stiglitz to take the floor.

Professor Joseph E. Stiglitz, Columbia University:
It is a real pleasure to be here. What I am going to try to do today is to talkabout some of the issues facing Japan, but I am going to try to approach it froma more theoretical perspective. There is a book that I have written with mycolleague at Columbia, Bruce Greenwald, called The New Paradigm of MonetaryEconomics that will be coming out fairly shortly. What I want to do today is tofirst talk about this new paradigm of monetary economics and then to comment onhow that new paradigm sheds some light on the particular problems that arefacing Japan today.

I am going to begin with a brief and fairly quick discussion of some of theproblems of traditional Keynesian macroeconomics. This goes back sometime, theseconcerns go back several decades. The problem was that it was recognized thatmuch of traditional Keynesian microeconomics was not based on micro-foundations.And in particular in the 1970s there was concern that the behavior could not bededuced from hypotheses of rational actors maximizing utility of profits.

A second aspect was that traditional Keynesian economics did not seem to addressthe problems of the day and made assumptions that seem inconsistent with whatwas widely observed. In particular, Keynesian economics as its traditionallyformulated focused on problems arising price rigidities. In the 1970s, theproblem was not price rigidity, the problem with inflation. And today, manycounties are worried about deflation. So obviously, the notion that there arerigid prices does not make any sense in a world in which we are worried aboutprices coming down year after year. And indeed, interestingly, even in the1930s, the assumption of price rigidities really did not make a lot of sense.Prices actually fell by about 30 percent during the Great Depression. And it isone of the great swindles that we persuade students for generations, the idea ofwage and price rigidities by never showing them the data. But in fact, pricesand wages fell down enormously and that means that we have to think about intimes of economic downturn, think about problems of downward price movements,that is to say deflation.

A second important problem with standard theory was that at the coremacroeconomics, of course, is monetary theory. But the theory of money that hasbeen the basis of much of macroeconomics does not make a lot of sense. Again, itis one of those great ideas that if you repeat it long enough and often enough,students come to believe it, but it really makes no sense, and it has made lessand less sense over time. The basic standard idea is that money is needed fortransactions. You need money to engage in transactions. And that that leads to atransaction's demand for money. The equilibrium interest rate is determined asthe intersection of the demand and supply for money. And that determines theinterest rate and the interest rate then determines investment, and there is achain of reasoning that goes along that way.

If you think about it for a minute, one recognizes that money is not needed formost for most transactions. Today, we almost never use money. Typically when Ivisit another country, I do not even change any money. I use my credit cardcompletely and that is true with most transactions today. Moreover, most moneyis interest bearing. In fact, the interest rate on money is simply the T- billrate minus a little transaction cost. I will show you a slide in a few minutesillustrating that, but the basic point is, that the interest rate that peopleget on their demand deposit which is what most money is, has nothing to do withthe transaction's demand for money. It is just determined by the transactioncost of converting T-bills into money. That means that the opportunity cost ofmoney, that is the difference between interest rate on-I use the term cashmanagement account, I should be careful because that saying a registered trademark of Merrill Lynch and in United States I would be hold before intellectualproperty for using a brand name, but I think you will understand what I mean.These are interest-bearing brokerage accounts that most people keep their moneyin today, most people with money keep their money in. That is to say people whodo not have any money do not keep their money there, but people with largedeposits use these kinds of brokerage accounts. So the opportunity cost of moneyis only the difference between interest rate on CMA accounts and T-bills andthat is simply determined by the transaction cost of CMA accounts and it isunrelated to economic activity.

Thirdly, most transactions are trades in assets and not directly related toincome generation. In fact, if you look at the volume of transactions, it is intrillions of dollars, and that is people buying and selling stocks andcurrencies, and that is to say not related to income generation, but that iswhat monetary theory is supposed to be about-not transactions but about incomegeneration. That would not be a problem if there were stable relationshipbetween income generating transactions and total transactions. But in fact, therelationship between the two was not stable over the business cycle.

And finally, the one of the basis of transactions demand for money was based onthe stability of velocity or the stability of the demand curve for money. But infact, it has exhibited enormous instability, which has led to the end ofmonetarism in most countries, fortunately.

This leads to a number of empirical puzzles and problems. One of them is thatthe standard theory puts a great deal of emphasis on the role of real interestrates, in determining economic activity. But as that first chart shows, realinterest rates for long periods of time have been almost constant. You cannotexplain anything by a constant. You cannot explain the business cycle by aconstant. So in fact, what one sees there is yes there are periodic changes butthey are not cyclical in nature. So it is very hard to use variability in realinterest rates as a central feature of a business cycle theory.

Secondly, if you turn to the next statement, there is a little evidence of thefact that real interest rates on investment in the United States. As I say,there are a large number of empirical studies, but in general, these studiessuggest that the dependence on investment on real interest rates is verylimited. And what we see in the United States, for instance, in the most recentrecession is that as the Fed has cut the nominal in the real interest ratesthere has been no effect on the investment. It has had some effect on consumerbehavior through refinancing a mortgage, but very little effect on theinvestment. Interestingly, a number of empirical studies suggest that there isconsiderable evidence of effects of nominal interest rates on the investment,which is of course, inconsistent with standard neo-classical theory which saysthat just real interests rate matter.

And part of explanation lies with the final observation on this chart, which isthat investment equations in which cash flow and net worth appear significant.This is an interesting episode in intellectual history. The first investmentequations that were done by Meyer and Kuh in the 1950s show cash flow effectsthat were significant. Then Modigliani and Miller wrote their paper andneo-classical theory grew and everybody said you cannot have cash flow effects,that is inconsistent with economic theory. So it became a heresy to have a cashflow affecting your investment equation. That continued until basically the1980s when the imperfect information theories developed, which said that cashflow and real balance effects are important. Then it became permissible to runregressions again with cash flow and real balance effects in. And lo and behold,what they found was what Meyer and Kuh found in the 1950s that those are verysignificant. So the standard investment equations basically of Paul Jorgenson donot work as well as alternate specifications.

There are host of other anomalies in questions. For instance, one of thestandard theories in macroeconomics is that inventories are supposed to bestabilizing. They should be a buffer. You should invest inventories in downturnsand reinvest in booms. In fact, if you look at the data, they are pro-cyclicalrather than anti-cyclical. A second example is that the experts often do notseem to increase as much as one would have expected after a large devaluation,large depreciations, which were seen very clearly in the East Asia crisis. Andfinally, there are problems with the movements of real product consumption wagesover the business cycle, which are either anomalous or not addressed by thestandard theory.

Two of the important questions which the standard theory does not really addressare why some shocks get amplified, which while the standard theory says that theeconomy has a number of buffers that should stabilize the shock rather thanamplify them. And finally, why the effect of some shocks seem so persistent.There are number of phenomena, reasons why you would expect that if you have ashock, one period negative, another period should be positive and the two shouldgo against each other. Therefore, you should find a great deal inter-temporalstability. But in fact, we see persistence in adverse shocks.

These problems on the empirical side have gone hand in hand with some advancesin economic theory, which have undermined much of the standard theory. And therewere two aspects of that that I want to mention. One related to the Nobel Prizein 2001 and the other one to the Nobel Prize in 2002. One of them focused onimperfect asymmetric information, which leads to imperfections in product, laborand capital markets. The others are the systematic irrationalities in behaviorwhich Kahnemann, Tversky and many others have identified, which have reallyundermine some of the basis of rational behavior. In particular, I will commentabout that in a few minutes, some of the implications. Anyway, to go back to thefundamental problem that was posed two, three decades ago, the fact that therewas the lack of micro-foundations for standard macroeconomics. There were twoalternative approaches. One was a theory of new classical and real businesscycle theories that were based on the standard microeconomics and tried to forman aggregate macroeconomics based on standard, neo-classical economics. Andsecond are a set of new Keynesian theories.

I first want to explain why the new classical and real business cycle theoriesfailed. Of course, I do not think they would agree with me on this, but I thinkthis is quite overwhelming, the failures. The fundamental problem, of course, isthat the models assumed away the problems that were to be explained. If you aregoing to try to explain unemployment, you cannot begin with the model thatassumes full employment. As Chairman of the Council of Economic Advisors, one ofthe problems I faced with, we had a higher macroeconomists and many of the verygood macroeconomists were coming out of places like the University of Chicagoand Minnesota. And all these economists believed that there was no such thing asunemployment. I had this image of hiring one of these and then having a meetingwith the President, and you have to remember President Clinton was elected onthe platform of jobs, jobs, jobs and he was very worried about unemployment. Ijust could imagine this meeting between this economist from the University ofChicago and President Clinton and President Clinton says, "I am very worriedabout unemployment" and this economist saying, "There is no such thing asunemployment. People are just enjoying leisure." And they might say, "Theydo not seem very happy about all this leisure. Leisure, you are supposed to beenjoying yourself." And he says, "That is a problem for the psychiatrist andpsychologist, not the problem for the economist." So I decided not to hire oneof the Chicago, Minnesota economists. Because I felt that he would fire meinstead and I would join the unemployed.

More broadly, it ignored mounting theory and evidence concerning imperfect,asymmetric information and irrationalities. This is really an important problembecause the basic methodology in the Chicago kind of approach is that therepresentative agent model. Now, think about a minute, all the work onasymmetric information on how capital markets have problems, based in asymmetricinformation, which I happen to think very important. If you have arepresentative agent, the only way you can have asymmetries of information, areif you have people with schizophrenia. That is to say, half their brain does notknow what the other half the brain knows. And that does not make any sense. Ifyou have only one person in the economy, you cannot have problems of asymmetricinformation. So the very basic methodology that underlay new classic economicsand real business cycle are the representative agent was doomed to failure ifyou think that one of the key problems is lack of imperfections of informationand asymmetry of information. It was just a methodology that could not work.Moreover, Shiller and others have emphasized that the stock markets exhibitirrational exuberance and pessimism, herd behavior. I think there is arecognition that this plays a very important part in the economy. AllanGreenspan, for instance, has emphasized the importance of irrational exuberanceand I think he today might talk about irrational pessimism.

Obviously, I think the problem of asymmetries of information are important andin fact, the host of corporate, accounting and banking scandals in the UnitedStates and Europe are related to imperfections of information and those scandalstoday are having, I think, significant macroeconomic effects. Moreover, if youlook in greater detail at the rational expectations model, you discover thatmost of the results have nothing, nothing to do with rational expectations, butdepend on perfect market assumptions. In fact, in some work that I did a numberof years ago with Peter Neary in Dublin, was that with market imperfections andrational expectations, the efficacy of government policies actually increased,not diminished. So rational expectations may actually make government policymore effective. What gave all the results of the rational expectation schools isthe assumption of market clearing. But of course, that was what you were tryingto analyze. Just to try to explain why that is so, think about the followingway:
one of the reasons that the government policies has limited efficacy has to dowith leakages-that if I increase incomes today, some of that is saved andleads to increased income tomorrow. But if people have rational expectations andthey realize that, and they realize their income tomorrow is going to be higher,that means they are going to consume more today. So there are theseinter-temporal feedbacks that actually increase the efficacy of policy if youhave strong rational expectations.

Finally, rested on a number of implausible assumptions such as the major sourceof disturbance and technology shocks, and the economy randomly becomes lessefficient. I think to me it was absolutely clear that those sets of approachesbased on trying to move from standard, perfect market models to an aggregatemacroeconomics was not going to succeed in explaining most importantmacroeconomic phenomena. So that leads one to try to think about alternativeformulations. And in fact there were two branches of what are sometimes callednew Keynesian economics. One is based on a more traditional rigid wages andprices, but as I said before, wages and prices are not rigid and many of thetheories like those of the new Chairman of the Council Economic Advisers, GregMankiw, are very unpersuasive and I hope he has better luck in his economicadvice than he does in this economic theory. The alternative approach is basedon debt deflation theories. Irvin Fisher, who was a Professor at Yale,emphasized the problems caused in the Great Depression by the fact that wagesand prices were falling and the fall decreases in asset prices. And in many wayswhat I am going to try to argue is this is the intellectual framework that oneought to be using in Japan and many other countries. This is a theory that Ihave developed with my colleague Bruce Greenwald and in particular it is basedon theories of asymmetric information and asymmetries in the speeds of priceadjustment. Obviously today, I can only hint at the broad outlines, but I wantto talk about a few of those.

The underlining idea is that we emphasize a great deal of imperfections ofcapital markets, and in particular, this leads to credit rationing of what wecall inequity rationing. That is to say, imperfections of capital markets thatlimit the use of equity markets in raising new funds, and therefore limits theability to spread risk. And that means that firms act in a risk averse way, italso means that firms' balance sheets matter, matter for production,investment, employment and all decisions and so do firm cash flows. It provides,in another words, explanation, for instance, the kinds of investment equations Idescribed earlier where real balance effects in cash flow effects do matter, aswell as, what is happening in the banking system. I will come to that in asecond. It also means that household and government balance sheets and cash flowalso matter.

One of the interesting aspects of this theory is that it focuses not only on thedemand side but also on the supply side. But it is very different from Reaganitesupply side economics. In other words, it focuses on some of the limitations onthe willingness and ability of firms to produce. In particular, since productionis risky, and risk cannot be fully divested, shocks to the economy can affectthe willingness and ability of firms to produce. It is especially relevant insmall open economies. The reason why-and this is one of the observations thatactually led us to the development of this theory-for small open economies,there should never been an aggregate demand problem. Simply by changing theirexchange rate, they should face a horizontal demand curve. They should be ableto sell as much as they want in the market. So if demand were the only problem,you would never see unemployment. Because just by adjusting exchange rate, youcan get as much demand as you wanted. The fact is that aggregate supply is asimportant as aggregate demand and when there is a shock to the economy, theability and willingness of firms to supply may be reduced. Particularlyimportant in this area is the supply of credit can be a critical constraint.Here in East Asia, one saw that in a very important way in the East Asia crisis.Because even as the country devalued, and their exchange rates fell, exportsincreased very little. And the reason was they could not get the working capitalthat they needed to increase their production. Actually Japan played a veryimportant role in the Miyazawa Initiative in providing some of the workingcapital that allowed the expansion of production that helped reunite theseeconomies.

One of the consequences of this is that demand and supply are very closelyintertwined. Demand shocks at one period have consequences for supplies insubsequent periods. Adverse effects on from balance sheets lead to lowerproduction, and adverse effects on bank balance sheets lead to lower creditsupply. So the important notion here is that shocks to demand in one periodaffect supply in the next period. And remember what I said earlier, one of thedifficult questions that one needs to explain is the persistence of the effectsof shocks. This theory then explains that in a very neat way. Because it says ifyou have a demand shock one period, it has effects on the ability andwillingness to both demand and supply in future periods. Finally, it alsoemphasizes the importance of redistributions. Redistributions, for instancecaused by large price changes, matter because of important non-linearities. Inother words, many of the changes that go on the economy are changes in prices.Say the price of oil goes up, that makes oil producers better off, oil consumersworse off. In a closed economy, in a standard economic theory, that would makeno difference because the gains to one would be offset by the lost to another.

You think of the global economy the same way. You have an oil price shock. Thegains to some should be offset by the losses to others. But once you recognizethat there are these real balance cash flow effects and they are non-linearities.The gains to one may not offset the losses to the other. Losses are felt morestrongly than the gains. And therefore, shocks can have a very strong negativeeffect. Again, let me give you an example where that was brought home veryforcefully. There were the oil price shocks that affected the United Statesnegatively and most other countries in the early 1970s, 1973 and 1979. Butinterestingly, in 1986 there was another oil price shock, in which the priceswent down. Now, standard theory said when the prices went up, you are worse off,when the prices go down you should be better off. But in fact, the US economywas hurt both times. That is because of these large distribution effects.

So leads to the new monetary paradigm which focuses not on base money, not onmoney that the transactions demand for money. It focuses on credit, credit asthe real engine in the economy. In fact, you can think of it as thegeneralization of loanable funds theory that was developed actually parallel tothe Keynesian theory by Robertson in Cambridge. But it is markedly differentfrom the old loanable funds theory because the key issue in the supply of creditis the problem of information, certifying who is credit worthy, monitoringcredit worthiness and that brings one to the institutions, the banks, whichprovide credit. Firms also provide credit. In our fully developed theory we lookat credit supply not only by banks but also by firms and by other institutionsin society. But the important point is that banks are specialized institutionsthat focus on the problems of information, ascertaining credit worthiness andmonetary enforcing loan contracts. One can view banks as firms that engage inthese credit services. But that entails risk bearing. And the reason why Icannot describe here in full detail, but one can separate out the informationservices from the risk bearing. Those are intimately intertwined and the resultof that is that as it provides credit information and provides credit, it has tobare risk. But the willingness in ability to perform this risk depends on bank'sbalance sheets. So that when the bank's balance sheets are adversely affected,the supply of credit gets adversely affected.

The theories focus then, on how shocks to the economy and policy-both macropolicy and regulatory policy-affect banks and others, including firm'sability and willingness to provide credit. One of the important aspects of thisis that there is often been this dichotomy that you think of regulatory policyas a micro-policy and central banks involved in macro policy. One of theimportant things that this emphasizes is that regulatory policy-bankregulations-have major macroeconomic effects. Again, to turn to the UnitedStates as an example, the reforms and regulations that were part of the 1989 USBanking Law were the major factors leading to the economic recession in 1990-91.One of the things that the Clinton administration did in 1993 to help theeconomy get out of the recession was to change the regulatory structure. So wevery much view regulations as part of macroeconomic policy. The problem was thateverybody recognized there was a problem of credit crunch. And that had to dowith the ability and willingness of banks to supply credit. So we had to thinkabout how we would affect the willingness and ability of banks to provide creditand that has to do with regulatory policy.

The theory pays special attention to bankruptcy, credit interlinkages amongfirms, which are as important as standard equilibrium product and factorinterlinkages. The new paradigm provides a framework for thinking aboutdeflation and alternative policy responses. Deflation and particularlyunexpected deflation leads to real balance sheet effects which can averselyaffect aggregate demand. In other words, what happens when you have deflation isthat you pay back more than you thought you were going to have to pay back inreal terms. You are worse off. The United States had a very important episodedeflation in the end of the 19th century and it was leading to some very seriousproblem for the US economy. Interestingly, the election of 1869 was fought theissues of monetary policy. It was the key issue in the election. The slogan ofthe Democratic candidate was that we shall not be crucified on the cross ofgold. And what that meant was the Democrats, who represented small farmers thatwere the debtors, said we have to increase the money supply. At the time, forcentral banks, the major method of increasing money supply was going from thegold standard to bimetallic gold and silver standard. That would have increasedthe money supply that would have undone the deflation. So you think of monetarypolicy as something that should be reserved to independent central banks, here ahundred and some years ago, it was the central issue of the political debate ofthat time.

This is in addition to the traditional real interest rate effects that many ofthe people have discussed here and elsewhere recently, when you have deflation,even when you have zero interest rates, nominal real interest rates can be quitehigh. In the Great Depression when the prices were falling ten percent a year,nominal interest rates were zero, real interest rates were 10 percent-nowonder economic activity was choked up. It is not that bad here in Japan, butthe important point is that when there is deflation there are these realinterest rates effects.

Globalization has lead to some deflationary bias in the global economy. Closerintegration could mean that there is deflationary contagion. Right now here inJapan, there is a lot of concern precisely about that issue, about whether theinexpensive supply of goods from China and deflation in China is leading tolower prices of inputs into Japan, and that is one of the structural factorsleading to a deflationary bias in Japan. Moreover, there are some otherstructural features that more competition is also leading to downward pressureson prices.

There is another important aspect that I not going to have much time to talkabout today, but I just wanted to note it because I think that it is one of themost important issues on the global architecture. We had a lot of discussionsfive years ago about reforming the global financial architecture. Unfortunately,it did not get at what I view as some of the most important issues. And this isone of them. The global reverse system means substantial global income is simply`buried' in the ground every year. What am I talking about? What I amtalking about is the fact that there are now more than $2 trillion of reservesand every year, something like several hundred billion dollars are added toreserves. What does that mean? That means that every year, several hundredbillion dollars of income are taken and put into the ground. Now the form weshould go with right now is US dollar bills. In the past, the deflationary biasassociated with the global reserve system was offset by the fact that manycountries ran loose monetary policies and this was offset by countries living bytheir means. But in today's global climate, that is no longer true. Everybodywants to have trade surplus. But you cannot have a trade surplus because one ofthe basic definitions is that sum of the trade surpluses have to equal sum ofthe trade deficits. So if China and Japan and few other countries are havingsurpluses and the rest of the world have deficits. But as they try to get rid ofthose deficits, it is like a hot potato as goes from one country to another. Andthe traditional way of getting rid of the deficit is trying to deflate youreconomy. That is, of course, what happened in Korea and East Asia. So it is asystemic problem in the global system.

Right now, we see this kind of mentality affecting Europe. At the same time, thestability pact in Europe means they have a limited ability to use expansionaryfiscal policy. Meanwhile, the central bank focusing exclusively on inflationmeans that they cannot use the monetary policy and Europe is, therefore, in whatyou might call a low income deflationary bias.

Now I want to come fairly quickly to the prescriptions. The theory suggests thatone ought to focus on the balance sheets. Shifting from deflation to inflationmay help the balance sheets, undoing the damage that deflation has done inincreasing the real value of debts. And the problems are particularly severe inJapan and other countries where there have been an asset bubble that breaks.Because when you have the asset bubble that breaks, it really destroys balancesheets. Moreover, shifting from deflation to inflation may lead to lower realinterest rates.

Now, there are three aspects of policy frameworks that I want to mention. Thefirst is moving from the deflation to inflation. The second is depreciation ofthe currency, a weaker yen. And the third I want to talk very briefly about isthe bank balance sheet because there are a lot of discussion of the problem ofnon-performing loans. In terms of-I should say depreciation, not devaluationbecause it is a quasi-flexible exchange rate, so I meant depreciation.Obviously, it will improve Japan's balance sheet, given its large creditorstatus.

The second point I want to emphasize, and this is quite important, the mindsetof central bankers around the world has been largely set by the problems of the1970s and 1980s when they were thinking about inflation, and that is what wastaught at graduate schools, that is what was taught in the economic courses.There is a real challenge for both central bankers and macroeconomists moregenerally, to begin to think about not an inflationary world, but deflationaryworld. And many of the things get turned upside down when you change yourmindset. For instance, one of the arguments against depreciation is that it isinflationary because the cost of your inputs goes up. And it is one of thereasons that IMF always is telling countries do not depreciate, do not devaluebecause they are always worried about inflation. They are fighting the war ofthe 1970s. That war is largely over. We want to go to the next war. In thatworld-the world of deflation-when the currency depreciates, it undoes thedeflation. And that is a good thing. So what was bad in the inflationary worldbecomes good in the deflationary world. And there are lots of examples of thiskind of change in mentality that one has to go through. And of course, thesebalance sheets effects, these anti-deflationary effects are in addition to thenormal trade benefits that results from depreciation.

Finally, let me talk just a minute about the issue of the bank balance sheetsand nonperforming loans. There is a lot of mystery and confusion in this area.One needs to separate out the issues of management from the issues of balancesheets. If you move nonperforming loans off the bank's balance sheets, movethem into a government corporation or some other corporations, and you pay thefull market value for them, you do not improve the balance sheet. It does notchange the bank's net worth. You solve the management problem, banks may ormay not be well suited for resoling these problems. They may not be well suitedfor dealing with nonperforming loans. But you do not solve their balance sheetproblems simply by moving the nonperforming loans off the balance sheet.

What does improve the bank balance sheet is if you take those nonperformingloans off the balance sheet, and you pay them as if they were performing loans.That is to say you subsidize the banks. But obviously, you can improve the bankbalance sheets by subsiding them in any way. So the real question is, is theagenda an agenda for subsidizing the banks or is it an agenda for management,which means, saying that they are not well equipped for restructuring loans.These two quite different agendas have been confused. It may well be that onewants to subsides the banks and I think there are some good arguments forrecapitalizing banking system, but if you do that, the government should getfull value for it. It should get an equity interest in the banks if itrecapitalizes them or has some other claims on the private banks. There is noreason why it should be giving these free gifts. But if you do not have thatkind of transparency, you are likely to make some very significant mistakes andit has some serious problems.

I also want to point out, and this is the final point, there are someunconventional ways which provide, by which one can recapitalize the bankingsystem which do not just move the problem of the, they do not, at the same time,worsen the government's balance sheets. In other words, there is a realproblem here. If you have a hole in the bank's balance sheet and you fill thatby government, you have just moved that from one place in the society toanother. And that may be a good thing to do, but you ought to recognize that youare not changing the underlining problems. You are just moving it around andthat is where the redistribution issues that I talked about before make adifference. The theory that I described says that those distribution issues areimportant. They should not be ignored. But you should recognize again what youare doing.

I am running out of time, I have a slide here that you can look at, trying toexplain why [break in tape] that I think one might want to think about, which istemporary consumption tax cut, may be more effective than income tax cuts for acouple reasons. First, the fact that it is temporary is more credible becausegiven the huge debt GDP ratio, it is more likely that those taxes will beincreased in the future, so that the temporary nature is credible. Whereas apermanent income tax cut is not credible. Secondly, by being temporary, youagree to inter-temporal substitution effect (i.e., induce people consume nowwhile there is a sale on consumption).

The key questions are following. I think there is an increasingly broad sense ofconsensus that one needs to reverse the deflation. And one needs the advantagesof depreciating the currency, the yen. So those two, there is a general, broad,increasing consensus on those two objectives. The problem is that those areendogenous variables. The rate of inflation or the rate of deflation is not amatter of government fiat and in the market economy the exchange rate is not amatter of government fiat. They are both endogenous variables. So the realquestion is and the hardest question is what government policies are there thatwill result in undoing the deflation and will result in a depreciation of thecurrency. There are number of policies which one can talk about and that havebeen discussed in Japan. I actually think that there is not one single panacea.One needs a whole set of policies because the problem is sufficiently severe. Itis going on for decade that one should actually have several policies. So inthis discussion today, I thought I would bring up one policy that is a littlebit unconventional and when I raise it, I almost worry that I might lose my cardas a credited economist because it is almost like heresy to say what is in thenext sentence.

Printing money, that is almost worst sin that economist commit. But one has tounderstand, again going back to the point I made before, when you are in aninflationary economy it is different from deflationary economy. In inflationaryeconomy, you should take away my card, my Ph.D. in inflationary economy. If Isay printing money, you should look at me and say where did he get his Ph.D.from? But in a deflationary economy, the things may be just the opposite. At thevery least, it seems to be an idea that is worth discussing. So the idea is tofinance some of the deficits by printing money. There is no evidence of adiscontinuity. That is to say, if some people say "if you start printingmoney, won't you have hyper inflation." That is the theory that says thatthe world is very discontinuous and certainly my observation about themoderation of both the Central Bank and the Ministry of Finance in Japan is thatonce they start printing money, they will not just turn the printing pressesfaster and faster. It will be very moderate. In fact, the real danger is theywill not do enough, not that they will do too much. So there is no evidence as Isay of discontinuity, a moderate amount will not set off rampant inflation andeconomic theory would say that there is, you could adjust the amount to get theright amount of inflation. It has a number of advantages over debt finance. Oneof them is that with debt finance, you have to keep ruling over the debt-everythree months, six months, year, five years. By issuing money, you do not haveto. It is permanently out there.

There is a second important advantage, that in most accounting frameworks, it isnot treated as part of government debt. One of the problems of Japan today isthat because it has run deficits year after year, the debt/GDP ratio now is thehighest among the G7 by a considerable amount. And that is beginning to havesome effect on rating agencies and one worries about a scenario in which we knowmarkets are irrational, we know markets can panic and one worries with opencapital markets that if the debt/GDP ratio gets too high, there could be thatkind of panic. If you ask the question, if your deficit/GDP ration is sevenpercent a year, in five years that is 35 percent. You add that to your already135 percent debt/GDP ratio, and the number keeps getting larger and larger. Soit is clear that the current strategy cannot continue and you have to thinkabout an alternative. One of the points is that this printing money can be usedto help recapitalize the banks. I talked about unconventional ways that wouldnot add to the debt burden of the government. This is one way, one of theseveral possible ways. In fact, this is a strategy which worked in Sweden in theGreat Depression.

There are some important lessons that even if the prescription works, it willnot address Japan's longer run problems. There are structural problems, aswell as aggregate demand problems. The two are intertwined. But I believe verystrongly, that one cannot affectively address the structural problems until oneaddress the aggregate demand problems. If you do not address the aggregatedemand problems, you resolve some nonperforming loans, you will get more innonperforming loans in another year or two years. And that, of course, has beenthe history in many countries.

I think there are opportunities for addressing some of the long-run productivityproblems, structural problems. For instance, one of the issues is that the levelof productivity increases in the service sector has not been matched at in themanufacturing sector. If you look around the world, the economies, there havebeen a major change in the structure of most of, many of the advance industrialeconomies, moving from manufacturing to the service sector. And economies thathave been most successful like the United States and UK have had enormousproductivity increases in the service sector. And yet, this is one of thesectors which is lagged behind in Japan. So that in some sense, that structuralproblem is one of the things that I think needs to be addressed. But thosestructural issues as I set repeat can best be addressed when the economy is in amore robust situation. Unless Japan does something about the continuing shortrun lack of aggregate demand, there is a real risk that the structural reformsactually exacerbate these long run problems as financial problems mount andinvestments in new technology wane.

Let me just come to some general conclusions. There is I think a new need for anew macro theory and a new paradigm for monetary economics. I think thedeficiencies in the standard Keynesian economics and the deficiencies in realbusiness cycle in rational expectations theories are sufficiently great thatthere is a need for alternative theory, one particularly adapted to the kind ofsituation that Japan faces with deflation. I think the new theory that I triedto outline here very briefly provides a better explanation of a host ofphenomena. Most importantly, the new theory actually provides insights into howto think about the variety of policy issues, approaches which are in many wayssignificantly different from those of either the standard Keynesian economicsand/or those of the rational expectations real business cycle theories. So Ihope in the discussions that we have, we can try to talk about the extent towhich this new theory provides insights into the situation in Japan. Thank you.

Professor Naoyuki Yoshino, Keio University:
Thank you very much Professor Stiglitz. Could you come back to the table here? Iwould like to ask….I would like to continue to speak in Japanese. I would liketo invite a Special Advisor to the Cabinet Kuroda and he will remain seated hereand speak with the panel.

Mr. Haruhiko Kuroda, Special Advisor to the Cabinet:
I would like to briefly provide comments to Professor Stiglitz's presentation.As you can understand from the presentation of Professor Stiglitz, ProfessorStiglitz is drawing theories from the realities in the economies and coming upwith policy proposals based on the economic theories. I am very much impressedwith this. Concerning the problems confronted by Japanese economy, he hastouched upon them. He has also proposed the policy proposals. Basicallyspeaking, I agree entirely with the theory in the policy recommendations ofProfessor Stiglitz. Therefore, my comments only have to do with the differencesin nuance, but I would like to make some points.

The first that I wish to make has to do with whether if we opt for the credit ormoney. In the banking sector, that is to say, central bank and commercial banks,the consolidated balance sheet, if we are looking into this balance sheet,basically speaking on the asset side, we see bank credits. That is say, creditprovided to the private sector and credit provided to the government sector.Also in terms of the credit extension, there could be loans and there also couldbe JGBs and corporate bond securities holdings. In any case, this is credit forthe bank. On the other hand, if you look at the liability side of the balancesheet of banks, it is mostly cash and deposit, M2 plus CD. So this is a broadmoney supply. Of course, equity also exists and on the asset side, there arephysical assets, but the means in the case of bank balance sheet on the assetside, we are seeing bank credit and on the liability side, we are seeing moneysupply. Therefore, non-bank credit, unless it is important, the credit theoryand money theory are almost identical. They are indifferent to each other.

Of course, having said that, in reality, as Professor Stiglitz has explained,the transmission mechanism is very different from credit to money. Therefore,macroeconomic effects may differ as a result of that. If that is the case, theremay be evident differences between the credit theory and money theory. In orderto prove that, non-bank credit must prove to be very important in the case ofJapan. And what about the situation here in Japan? As Professor Stiglitz haspointed out, most of the payments are being made by credit card and in terms offunding of companies is that trade credit or is that a funding through thecapital market? That is a very important question that we must ask ourselves.According to my impression, perhaps not to the degree of the United States,non-bank credit is not as important in Japan. However, credit cards havepenetrated quite widely in Japan and the capital market is becoming increasinglyimportant in the case of Japan. So credit theory is much more relevant andeffective. So we have to look into the empirical studies of the Japanese economyand I think we must ask economists, academics to look into this. Next slideplease.

Professor Stiglitz has spoken about unexpected deflation, that it is much moreharmful. I agree with him entirely on that note. In the case of Japanesecorporate and government debts, of course this is colossal, both in the cases ofthe corporate sector and government sector. Over the past 20 and 30 years, wehave seen continuous accumulation of such debt. So such debt were made beforepersistent deflation was expected. Over the recent few years, of course thedeflation has been expected and the short-term interest rate is now standing atzero, and even long-term rate is one percent or two percent, standing at a quitelow level. Therefore, there are funds transactions based on expected deflationbut when it comes to debt in the government sector and also the mortgage loanfor individuals is a major problem. Of course, mortgage loans last for 30 yearsand the housing price used to go up and people were expecting the housing priceand asset price do go up. When there was still inflation, the consumers havedrawn on their mortgage loans. However, we are now suffering from deflationaryeconomy and therefore, this is a major pressure on the household account. Thecurrent deflation in any case is seriously damaging the corporate governmenthousehold debts. This was unexpected deflation and the current deflation isseriously damaging to Japanese economy. I agree with Professor Stiglitz on hisanalysis. Next slide please.

The third point is something that was not raised by Professor Stiglitz, but ithas been debated here in Japan so I would like to briefly go over the point.Some Japanese economists are arguing that open market operations, if they areconducted in large volumes in the long end of the market, such as 10-year JGBs,is really only a combination of the conventional monetary policy and debtmanagement for the BOJ to be conducting such monetary policy and debt managementat the same time. So for the BOJ to engage in large volumes of the open marketoperations in the long end of the market is questionable if there is a need.Above the yield curve, if there is a need for such market operations, as theGovernment of Japan, it should be engaged in such market operations. My positionis to oppose such arguments because this kind of bills only doctrine many, manyyears ago, the US FRB was involved in debts but this was more than 50 years andyou have given this up 50 years ago. And as far as BOJ is concerned, over thepast years what has been engaged in is open market operations of the JGBs, andtherefore, there is no reason for the BOJ to go back to the bills-only doctrine,but for a broad range of assets, not only ten-year JGBs but other types ofassets, the BOJ should engage, should be very bold in conducting open marketoperations and this is necessary to overcome deflation. So I just wanted toshare with you some of the debates here in Japan. Next slide please.

So what about monetarization versus money financing? Let me explain what I mean.As you might be aware, a substantial portion, maybe about 40 percent ofgovernment expenditures is debt-financed. That is to say, the government isissuing JGBs to finance government expenditure and therefore, there is really asurplus of JGBs and 1.2 trillion yen of JGBs is being purchased by the BOJ on amonthly basis and therefore, the BOJ is monetarizing such government debt.

Monetarization does not change the amount to debt of the government but it doesreduce the net debt service cost. The BOJ holdings of JGBs, that is to say thegovernment is paying the interest to the BOJ but the BOJ profits will increaseand the BOJ in the form of incorporation tax or in terms of other payments, ifthe BOJ will pay this amount back to the government, then this will have theeffect of reducing the net debt service cost. However, the total amount of debtof the government will not be reduced at all.

Professor Stiglitz is suggesting that the government, itself, should engage indirect money financing, which would mean that there would be no further increasein the amount of debt and the debt service cost can also be saved. So if thegovernment engages in direct money financing, this would reduce both the totaldebt and the debt service cost. I do not know whether the government has theauthority to engage in such direct money finance and the current mechanism isbeing set up for the BOJ to play this part, I do not think it will be possiblefor the Government of Japan to engage in that kind of direct money financing butI felt this proposal was unique and it was a very interesting idea. As far as Iknow, in the 19th century in the United States, there were many periods in thehistory of the United States where there was no central bank. And, as ProfessorStiglitz has pointed out, the gold or silver standard used to exist in theUnited States but if the US government to be issuing greenbacks not backed up bythe gold and silver standard there were time periods in which there were largecirculations of such government-issued notes and therefore the proposal ofProfessor Stiglitz, of course you have drawn upon the example of Sweden and alsothere is the example of the United States in the 19th century. Of course, thiswould be quite controversial and as far as I am concerned, I feel there are somesteps that could be taken before going to that step, for larger purchases ofJGBs to be made by the BOJ and if there is monetarization, the debt-service costcan also be saved. I think that avenue is much more realistic but the proposalmade by Professor Stiglitz I find very unique and very interesting.

Last but not least, this is the fifth slide. Of course, monetary expansion withother variables are constant, may lead to exchange rate depreciation asProfessor Stiglitz has pointed out. This will help to reduce and alleviatedeflation, which would be desirable. Having said that, it could be that directintervention could be made in a foreign exchange market so that the yen could bedirectly depreciated and this might be more effective; however, that will inviteinternational concerns. I am saying that because if there is direct interventionin the foreign exchange market, in terms of stability of the exchange rate,intervention is to be made in the exchange market in order that the exchangerate reflects the fundamentals in a better way. But direct intervention in theexchange market, in order to surmount deflation is considered to be exceptionaland is not widely recognized. Therefore, it might be difficult to directlyintervene in the exchange market to overcome deflation and therefore, asProfessor Stiglitz has also suggested, we would like to place emphasis on widerfinancial markets, the long term as well as the short term financial moneymarkets and also capital markets.

If we could pursue monetary policies, we should exert influence on the widerfinancial markets and should be emphasized rather than leading to yendepreciation through direct intervention of the foreign exchange rate and tryingto overcome deflation as a result of that. Various monetary expansion as alsosuggested by Professor Stiglitz would be more desirable in my eyes.

I did present some of my comments, having listened to the presentation byProfessor Stiglitz. But, basically speaking, I am very much attracted to theseries presented by Professor Stiglitz. I do hope that such series can beapplied here in Japan so that there can be empirical studies as well as policiesrecommendations made.

Professor Naoyuki Yoshino, Keio University:
Thank you very much. Now I would like to take this opportunity to offer mycomments. You should have handout number 4. The slide is prepared in English andI would like to start commenting on the paper. As Professor Stiglitz mentioned,the Japanese economy and Asian economies are so much concentrated on bankchannels rather than other channels. Especially the bank channel, it comes fromboth sides. One is from household sides and most of our household assets areeither deposited in banks or post offices. And how to diversify household assetsis also important. In general, Japanese people have a lack of management.Especially, during the late 1980s, I think politics have saved Japanesehouseholds not to face risk and 100 percent of deposits were fully guaranteed.

If we come to number two, deposits were fully guaranteed until last year andeven now, liquidity deposits are fully guaranteed. However, banks are facingrisks. Their non-performing loans will definitely rise and banks are verycautious so they hold lots of government bonds rather than making loans. Thennumber three, how to establish alternative money flow, rather than throughbanks, is also maybe important in Japan. Among OECD countries, Japan is the onlycountry in which households still make deposits of about 60 percent of allfinancial assets.

Number four is the devaluation or depreciation of the currency. I thinkProfessor Stiglitz mentioned, we are very close to China and their exchange rateis fixed and furthermore, their exchange is very much favored for their exports.In those circumstances, probably our exchange rate policy may be different. Ifall the countries are moving in free-flow or flexible rates, then that policymay be good. The next country is using a very different exchange rate policy sothat in those cases, a traditional textbook may not apply.

Number five, real interest rate versus nominal interest rate. As ProfessorStiglitz mentioned, real interest rate is insensitive to investment. Thatapplies to Japan too, especially when Japan is facing to lower interest rates,domestic rates are always insensitive. On the other hand, there is an asymmetry.If our central bank raises its interest rate, then that can reduce investment.So in the Japanese economic model, asymmetry exists in lowering rates versusincreasing interest rates. So in those cases, a huge money supply and at thesame time, lower interest rates cannot encourage domestic investment. In thosecircumstances, what is the best policy? Furthermore, utilizing fiscal policy isineffective, so our economy is facing a vertical IS-Curve. And we can shift thatIS-Curve to the right. I think that is the problem we are facing.

The next page, short-run macro policies. We have monetary and fiscal policies.Base money is increasing a lot, maybe 25 percent or 20 percent. However, moneysupply is growing much, much lower, due to the malfunctioning of the bankingsector. I think when we are talking about monetary policy, we like to talk aboutmoney supply in base money sense versus money supply in M2 + CD. Still, manytraditional fiscal polices are implemented by politics. And number two, as youmentioned, we have huge budget deficits, but nobody talks about how to stopspending. Many people are concerned about budget deficits but nobody actuallytalks about how to cut them, especially, the allocation of fiscal spending.Because of the Japanese election system, regions are very important and ruralareas still hope for fiscal spending and also elderly people need their pensionsand other benefits. We are facing an aging society.

Number three, you also talked about a lack of competitiveness of the servicesector, including the financing sector of the Japanese economy. I think that isvery much true; however, it may be somewhat related to English language skill.Because in manufacturing, we do not need any language. As long as we produce agood car for a low price, everybody understands. But in the finance and servicesector, we have to explain, so that as you said, we should shift frommanufacturing to service, but probably there is some disadvantage or we have toimprove our English-speaking knowledge and so on. Probably, as you are Englishspeaking you do not realize these issues. Most of the students spend huge amountof time for their English studies in high school and junior high school.

And number four, innovation in the financial sector. You touched upon theinnovation of technological progress. But, when I listened to many people in thebanking sector, basic skills and equipment are not lacking in Japan. Probablywhat is lacking is experience or new ideas, or management skills. So, probablyif we could improve those points, we may be able to catch up.

The next page, page three. Huge budget deficit, actually our deficit to budgetGDP ration is 141 percent rather than 135 percent. We cannot widen this anyfurther. Number two, that also causes some clouding out effect. In effect, JGBsare mainly held by domestic financial institutions. Then, is there any short-runfiscal policy possible? I do not think there is any room for fiscal policy rightnow in Japan.

And number two, you are suggesting some temporary tax cut or temporary incometax cut. Some empirical studies show that the late 1990s, Japan implemented anincome tax cut. That caused some widening of income inequality in Japan.Furthermore, income tax cut increased savings rather than increasingconsumption. Furthermore, in 1997, the consumption tax rate was increased fromthree percent to five percent. What happened in that time, was lots of purchasesof housing and automobiles were shifted before the consumption tax was raised.And of course, other consumer products remained the same, but some of theconsumption shifted before the consumption tax hike. So even if there is sometemporary consumption tax cuts, then once it has been raised, the consumptionmay drastically decline. So in other words, our experience in inter-consumptiontax hikes from three percent to five percent and I look at about five years ofconsumption patterns. Overall, there is not any change. Of course, there is ashift from the pre-high consumption tax period, there is a shift but overall, inthe Japanese case, there was no change. So even if there is a temporary tax cut,I am afraid it may not be so effective in the case of Japan.

The aging population is another problem for Japan. And number four, we stillhave lots of structural reforms, especially in agricultural and construction andso on. Since these sectors are protected and they are not facing thedetermination of supply and demand for the exchange rate determination, so thatcauses the yen to still appreciate. Probably, one more policy may be to openmuch further for the restricted sectors. That may accelerate the depreciation ofthe yen.

The last page is some empirical analysis about what are the explanatoryvariables or cause of Japanese inflation and deflation. These are the timeseries data by using the neural network econometric method. Number six,aggregate demand, has the highest reason why we are facing either inflation ordeflation. And number eight, import price, which is related to exchange rate, isnot so high. We are facing lots of cheap products from China but even in the olddays, we are facing other products. There is not so much impact on the importprice but rather the most important factor is aggregate demand shock. The impactof fiscal policy has declined very sharply and monetary policy is the secondone. And expectations are not so large because our economy has shifted frominflation to deflation, so the expectations are not so high. Again the importantfactors that explain from both aggregate supply and aggregate demand side arethat the demand components are the major factor.

The last one is a very minor comment about deflation and inflation, or the priceof land. Whenever we are facing some economic conditions, everybody worriesabout the same things. For example, in the late 1980s we were talking about highland prices being a problem. Once land prices go down, our economy will becomevery strong and so on. Then now, the fallen land price is a problem. Then wewere affected by high inflation in the 1970s and 1980s. We were talking aboutonce we stop inflation, we will improve everything. And, the same thing can beapplied even now. I think we still need structural reforms and other policyreforms to stop deflation. Thank you very much.

Professor Yoshino:
From now, I would like to open the floor to questions and comments. Since thisis the fourth meeting of the Foreign Exchange and Other Transactions Subcouncil,I do apologize that we have most of the distinguished guests in the bench andnot in the front row. Mr. Nanjo, please.

Mr. Nanjo:
Professor Stiglitz, when you were senior economist with the World Bank, youtraveled frequently to Japan and I was sometimes impolite in addressing yourquestions, but you were always smiling and very nice in coming up with responsesand I hope the same will be repeated today. I have two questions. Concerningincreasing printing money, as Mr. Kuroda has also pointed out, in the case ofJapan, the BOJ has a delegated and specialized function of doing this and it isdifficult for the Japanese government to overtake this. What the Yomiuri Shimbunis preaching is perhaps that non-bearer, non-interest JGB should be issued.Perhaps, this can be conducted on a temporary basis to respond to an emergencyand if the volume can be rolled-over, there would be no interest and this wouldbe permanent and there would also be some modernization. I think this would havethe same effect as the government going ahead to printing money. So may I inviteyour view on this suggestion?

The other question is Professor Stiglitz you have been persistently arguing foryen depreciation and theoretically speaking, yen depreciation is very muchnecessary and I think a large number of Japanese cater to that view. Mynewspaper also caters to that view. So how can we accomplish the yendepreciation? That is the crucial point. I agree with what Mr. Kuroda has said.Realistically speaking, smoothing operation, that is something that we have beenengaged in and this is possible to a certain extent. If there is to be a certainpolicy objective for the yen exchange rate to be induced to a certain direction,is nearly impossible in today's world. So trillions of dollars are beingtraded in the forex market and the degree of the intervention is quite limited.Unless we can have consent from the US government, we would not be able torespond at all. In the cased of the Plaza Accord, there was a consensus thatthere has to be adjustment but in terms of today as well as in terms of thefuture, I think this is immediately before the elections, President Bush of theIraqi War would not want to repeat the failure of his father Bush, Sr. and hasto try to improve the economy of the United States. So, a strong dollar may besomething that is preached but unless there is a cheap dollar, export relatedindustries in the United States will not vote for Bush in the next election.Therefore, we will not be able to forge a consensus of needing a strong dollarand GNP. So, even if we say that yen depreciation is desirable, in reality, thisis an impossibility today. But, is there a possibility for us to be able torealize yen depreciation under the current circumstances? That is the question Iwish to raise. And if you want to realize the yen depreciation, what are thepolicy instruments to realize this?

Professor Yoshino:
Perhaps, we can entertain several questions first. Professor Iwata?

Mr. Iwata, Deputy Governor of the Bank of Japan:
I would like to make two comments in English, following the suit of ProfessorYoshino. The first comment is relating to globalization. Page 17 mentions thatthe global reserve systems means substantial global income is simply buried inthe reserve every year. I was wondering whether this is really true or not. Ifthis is true, how can we correct this deflationary bias? In my view, the currentreaction of Asian countries, including Japan, accumulating massive amounts offoreign reserves in dollars, this is just a kind of accumulation of deposits andmoney by the household sector and non-financial corporate sector. The householdsector has ample deposit and financial assets, while the corporate sector alsoaccumulates free cash flow but which does not lead to real business investments.This is a big problem; this is a hoarding of money due to the proceedingdeflationary tendency in our economy. You are mentioning that this globalreserve system is a similar phenomenon, that is hoarding of money by monetaryauthorities is actually taking place.

But what I observe, I think is as follows; my interpretation is that thisaccumulation of foreign reserves is coming from the kind of protection policy ofexchange rate against appreciation. Japan and Asian countries want to avoidsharp appreciation of currency and make intervention on the foreign exchangemarket and accumulate the foreign reserves. Then your hypothesis is translatedinto this kind of proposition, whether dollar depreciation in effective terms orthe dollar appreciation in effective terms leads to deflation or inflation. Ifind some bias in your proposition as you are American, and you prefer thedepreciation of the dollar, which would lead to worldwide boom. That is mysuspicion. But, anyway, that is my first comment.

My second comment is on printing money by the government, the Ministry ofFinance. This is a very serious issue for the Bank of Japan. I have now justserved for one month as the Deputy Governor of the Bank of Japan and I mustprotect my position and say something on this recommendation. As ProfessorStiglitz argued, we can learn a lot from the experience and lessons in thecourse of the 1930s in Sweden, the United States and Japan. In Japan during the1930s, actually the finance minister Mr. Takahashi adopted this quiteexpansionary government expenditures financed by direct purchases of theGovernment Bonds by BOJ. That was a very strong policy, that money financedexpenditures and which led to inflation and loss of control over governmentexpenditures in the later period under the military government. So the currentfinance law prohibits the direct purchase of government bonds by the BOJ. Ithink this is quite an important point in a democratic society-which part ofgovernment including the central bank should play its own role- notably theindependence of monetary policy of the central bank.

Really, in my view, the independence originates from this lesson. If thegovernment has the freedom to print money, at the same time to control tax andexpenditures, there would be no control on the expenditures, that is the crucialissue and this leads to the problem of integrity of money-how to maintain theintegrity of money. If you look at the case of Argentina, you see localgovernments issues on money and what was the consequence of this. That impairsthe credibility of monetary policy and also the central bank. In Japan,historically speaking, we have in the Meiji and Taisho eras, seven times thegovernment actually printed money but the result was simply absorption by thecentral bank, replaced by the issuance of money. It created fake money and onlyconfusion in the financial market. So, this is not such a good recommendation-todirectly print money to finance expenditures by the government itself.

And a smaller point is seigniorage gain, or the distribution of seignioragegain. Now the current system, the BOJ has seigniorage gain, which should betransmitted to the government, but a certain portion of seigniorage gain is usedby the monetary policy and macro prudential policy carried out by the Bank ofJapan. Recently the BOJ moved to macro prudential policy, including the measuresto buy equity from banks, because Japanese banks hold large amounts of stocks.In order to stabilize the system, we embarked on this marcro prudential policy,buying this equity from banks. In order to carry out this operation we aretaking a real risk, or the case of asset-backed commercial paper, this is justlike a direct purchase of private bonds by the central bank. That implies creditrisk we are now taking. So, we should prepare against this risk; we should havesome more reserves. This is a subtle issue but it is also important-which partof the government or central bank accumulates seigniorage gain. Thank you.

Professor Yoshino:
Thank you very much. Dr. Stiglitz, could you answer some of those questions? Andof course we will ask more questions from the floor later.

Professor Stiglitz:
Can I begin by commenting on the first two? Let me first comment on Mr. Kuroda'scomments on credit versus money. The fact that the balance sheet of the bankingsector has money supply basically equal to the credit is one of the reasons whythere has been such longstanding confusion in this area, because if you do aregression, the two numbers are going to be the same and it is very hard toidentify what is the driving force. The fear that we are emphasizing is that itis really the credit supply that one wants to focus on and that the creditsupply that for instance increases in the base money may not translate directlyinto credit supply, which of course one of the key issues facing Japan today,that the monetary authorities control base money but they do not controldirectly the credit supply. In the end the two may equal but it is what do youcontrol. Similarly, it encourages the perspective of focusing on the creditsupply, encourages one to focus on variables like the spread the between T-billrate and the rate at which firms are borrowing. And that spread can changeenormously.

And example, where in the United States I think a mistake was made by focusingon the T-bill rate rather than on the credit numbers, was in November 2000, whenthe Fed tightened the money supply, worried that there would be inflationarypressures, just as the economy was going down. Now forget about theirforecasting problems. The fundamental problem was that if they had look at whatwas happening to the interest rates that firms were paying, those has alreadyshot up. What happened was in a way, the interest rate that is relevant for theeconomy had already increased so that when the Fed increased the T-bill rate, itincreased it even more and that caused a real problem. So this says that at keypoints in the economy, there could be a change in the spreads and it encouragesyou to look at those numbers, that again are an endogamous variable.

The second point I want to comment on has to do with the observation of thepoint between monetary policy versus debt management. I think here the keyobservation is that the objective of monetary policy should be macroeconomicstabilization, which means demand to maintain the economy at full employment.Now, whatever instruments do that should be viewed as the right instruments, sothis doctrine of bills only is a doctrine from the 19th century that has nothingto do with modern central banking so I would want to encourage a broad view onwhat are the appropriate set of instruments. In that respect, if the interestrate that is relevant for investment is the medium to long-term interest rate,if intervention can get the medium to long-term interest rates downs, then thatis an appropriate policy, so I think one should not take a doctrinary view onthat.

Let me comment on the intervention in the exchange rate versus financial marketsand the difficulty that was also raised in the question about how do youdepreciate the currency. As I pointed out and observed, it is an endogenousvariable. And what are the instruments? Well, as pointed out, there are a numberof instruments that can affect the exchange rate. For instance, if one cansucceed in convincing the market that one is going to undo deflation and move itto inflation, that may lead to a depreciation of the currency. So any policythat succeeds in moving from deflationary to inflationary may have that effect.Some people argue for inflation-targeting as an instrument. If that is able toget inflationary expectations, that makes sense. I am not convinced that is thecase and I do think that the exclusive focus on inflation, which is the case inmany countries following inflation targeting, is wrong. My own view is that oneought to use monetary policy to pursue all the major macroeconomic objectives-asin the United States, employment growth and inflation.

If one can use that as an instrument at the current time to change expectations,that seems to me something that one ought to consider. Another example isexchange rate intervention and this deals with a question that was raised iswhat will the political economy, what will the response of other countries be tothat, and in particular, the United States. Now, different administrations havetheir strengths and weaknesses. One of the benefits of having an administrationthat at least in principle is committed to market forces is that they do notbelieve in exchange rate intervention. Now, whether that policy will change isanother questions but certainly, in terms of the official stance, it is one ofnon-intervention.

There is a very difficult issue for all the countries in the world, which is wedo not have a world in which we have flexible exchange rates. China has fixedits exchange rates in terms of dollars. The dollar is decreasing in valuerelative to the euro. That means in terms of other exchange rates, China isdevaluing. If a major neighbor is devaluing, the question is what should yourexchange policy be. It is not clear. What is clear is that it presents aproblem. Part of Argentina's problem was that it had a fixed exchange rate,when its neighbor, Brazil, had a flexible exchange rate. This incongruence ofexchange rate regimes is a challenge to the global economy. One of the importantthings to recognize is that sometimes one can characterize policies in differentways. Korea characterized its policy in the aftermath of the East Asia crisis asone in which it wanted to accumulate reserves given the instability that wasclearly manifest in the global economy. Accumulating reserves also happened tohave a side effect of keeping the wan at a very low exchange rate.

Now, Japan could also decide that it is appropriate for reserve management toincrease the size of reserves. It would be a prudential policy that might havesome side effects on exchange rates but that would be just an accident. Andthere are other examples like that. I think argument for exchange rateintervention-I do not want to use that word, what I just talked about-isthat by being more targeted, the effects may be larger and the exchange rate isthe variable that you want to in some sense affect, it has broad macro andeconomic consequences and therefore, one might argue that having a target mightbe more advantageous.

Let me just make a couple comments on Professor Yoshino's remarks. The firstis I think he asked a very good question. He said what besides monetary andfiscal policies, what are the other policies that one can use. I think that isan important issue. One wants to take a very broad perspective on economicpolicy. Three that come to mind are one, the policy of increasing reserves giventhe global instability. I think that is one policy. Secondly, regulatory policyon the part of the banking system is something that I emphasized. And thirdly,there can be direct credit policies on the part of government. In the UnitedStates, governments in recent years has been involved in 25 percent of alllending activities. You talk about a free market economy, but that is only therhetoric. The reality is that through government guarantees and governmentlending, about 25 percent of all loans/credits when through the federalgovernment, had federal government involvement. Clearly, Japan could have a moreactive credit policy, for instance, for small and medium sized enterprises. Thatis another example of a policy.

The second important point that I think Professor Yoshino raised is thatdeficits by themselves are not necessarily the best way of measuring the impactof the government. Let me explain what I mean. The United States-and I willtalk it from the perspective of the United States-the Bush administration hasachieved enormous deficits. It is an achievement that I think no one could haveimagined. The turnaround in a ten-year window, in two years has been $5 trillionand he proposed another $2 trillion deficit. But, this $5 trillion has had verylittle stimulus. He designed so that all the money would go to the wealthypeople and it would not stimulate the economy. He worked very hard to getsomething that was so badly designed but he succeeded.

It is important to think about how the money is spent. In some ways, spendingmoney has more stimulus effects than others. We call it "bang for the buck."Sometimes I joke that in mathematics we always talk about making a mistake onwhether you maximize or minimize. He minimized the "bang for the buck"rather than maximized the "bang for the buck." So it is important to thinkabout how you spend the money. One of the things that I think one ought to thinkabout is are there ways of spending money that addresses the structural problemsthat Japan faces, because I think everybody recognizes that there is interplaybetween aggregate demand problems and structural problems. Money does help solvestructural problems so using the deficit spending for things that would increasefor instance private sector investment might lead to private sector investmentand shift the IES curve. So that would be an example of that kind ofexpenditure.

The third point that Professor Yoshino made and which I want to comment onbriefly is that he is perfectly right that a temporary sales tax cut shiftsconsumption from one period into another. In fact, that is what they are called:
inter-temporal substitution. You substitute from one period to another. Thecritical issue here is is there a threshold level in which you can do thisenough to restart the economy? I think that is really the debate; it is arecognition that this is not a permanent policy. This is a policy, is theresomething that can pump prime the economy to get it started it again? That is anopen question in my own mind.

I think a number of people have pointed out to me the institutional framework ofJapan does not make it easy for the government to print money. Obviously, thereare two answers to that. Well, one of them is to change the institutionalstructure. Everything is difficult and as the IMF sometimes say, you have toface painful choices, and this may be one of the painful choices, giving thegovernment the right to print money-particularly painful for the central bank.

I think the issue that was raised about the institutional design, theindependence of the central bank, the dangers of governments printing moneyrampantly leading to inflation are all things that are important questions. Itried to anticipate some of the those questions earlier by saying for instance,even with or without independence, I do not expect the Ministry of Finance tostart rolling the printing presses today without restraint. It just does notseem to me a likely outcome. More generally, if you look around the work today,there is very little evidence that an independent central bank leads to betterperformance. I know central banks do not like to hear this but in terms of thethings that really count, in terms of things like growth, unemployment, tradeoffeven in the Phillips Curve between what is called the sacrifice ratio-independencedoes not make much difference. The one thing that independence does have, it hasa small effect on, independent central banks focusing on inflation, do succeedin getting lower inflation by a little bit. But they do not foster economicgrowth, lower unemployment, any other variable that is important.

I think the idea of the independence of central banks has been oversold,particularly by the central banks. I also think that if one worries about theMinistry of Finance being reckless and printing money without regard, there aretwo points. First of all, some of these problems that have occurred, haveoccurred in societies that have not been democratic because in democraticsocieties, that kind of rampant inflation is checked at the polls. The citizensdo not like rampant inflation. I have certain confidence in democratic processesand believe that it is a check, and you see that in many countries that aredemocratic, that that is not much of a problem.

Second, even if you do not trust democracy very much, you can put a constraint.You can say the government can print money but it could only do it by apercentage that is equal to the unemployment rate or the deflation rate. Thereare ways of checking that provides constraints and given the nervousness thatsome people express about the recklessness of the government I can understandputting those constraints on.

Finally, I think the observation was made that if you could not do that, issuinggovernment notes and calling them bonds is close to a substitute. That is tosay, non-interest bearing perpetuities are equivalent to money. So one couldissue a non-interest bearing perpetuity and it could circulate just like money.I would not call it fake money. I would just call it alternative money. So thatcould address the problem.

Now there was a very interesting question about the issue I raise in my talkabout the global reserve system. Let me just very, very quickly touch on that.It is an old issue and in fact it was an issue that Keynes worried about at thetime the IMF was set up. It was an issue that was brought to the fore by thegold standard system, so it is a real issue and it has been a long-standingissue. At the time that the IMF was set up, Keynes actually proposed that therebe a tax on countries having surpluses. At the time, much like there is today,there was a worry about insufficient global aggregate demand. His view-and Ithink the was correct-was that if countries are putting money into reservesevery year, if they are not spending the money it would lead to a globalinsufficiency of aggregate demand. Building up reserves is like not spendingmoney and leading to an insufficiency of aggregate demand. So he proposed a taxon countries that had surpluses.

There is an alternative mechanism that I talked about and George Soros hastalked about, a few other people have talked about, and Keynes has talked about,and that is the creation of global greenbacks. I do not mind, greenbacks,yellowbacks, whatever. I am sorry to use the term greenback. But, global SDRs.The idea would be that every year you would automatically create these newspecial drawing rights that would be equal in magnitude roughly to the amountthat is buried in the ground in reserves. What would be interesting about thisis that these SDRs could be used to help finance global public goods-development,global environmental public goods, global health public goods-and so theywould be issued to developing countries that would widely end up down in people'sreserves. But it would be like an increase in the global money supply and offsetthe deflationary bias in the global system.

Professor Yoshino:
We would like to entertain further questions. Mr. Kan and Mr. Kumai please.

Mr. Kan:
Japanese policy makers have asked China to devalue its currency. I would like toask Professor Stiglitz if you would agree with this idea that a stronger Chinesecurrency would solve Japan's deflation problem? Actually, the numberspresented by Professor Yoshino show that the effect would be quite marginal.

Mr. Kumai:
Thank you very much for giving me this opportunity to speak today. I have twopoints. The first point is related to the question raised by Mr. Kan right now.In the case of Japan, there is mounting interest vis-à-vis China and one of themajor causes of Japanese deflation probably rests in China, according to oneargument. In the future, in the long run, how will China develop? This will, ofcourse, exert major influence on Japan as well as the United States. From arealistic view, Professor Stiglitz, you have formulated a theory incorporatingalso the aspect of China and the future development of China. What will be itseffect on the development of your theory in the future? I would like to inviteyour comment on this? That is my first question?

The second question also relates to something you have partially responded to. Iagree entirely with your recommendation that yen depreciation should be inducedand I believe that is perfectly possible, but within the domestic circles inJapan there are four major criticisms vis-à-vis that policy. The first critiqueis something that you have also referred to and whether the United States andJapan will allow to do that but I think you have already responded to this.

The second question is, of course, the Japanese government may have less controlin terms of moving the exchange rate. If there is 10 percent yen depreciation,there should be 0.6 percent positive effect on Japanese GDP, but Japanese exportcompetitiveness has deteriorated in the high technology industry and reverseimports are increasing and perhaps the effect of yen depreciation may not be aslarge as initially expected. That is the second point.

The third point is if yen depreciation is to be induced, this may lead tocapital flight and this may be a triple flight type of situation. For instance,banks have a large debt and a few trillion of yen in losses would be suffered.That is the third argument I wish to present to you.

The fourth factor is that of a more fundamental nature. If we are to induce yendepreciation, it means that Japan would maintain the current industrialstructure and according to you, Japan should try to recover it economy and thentackle structural reform, but this might have a more hazards or a tendency toprocrastinate fundamental solutions. So even if there should be structuralreform following the economic recovery, perhaps that may not be possible, so wehave to induce yen depreciation at the same time we are grappling withstructural reform. That is the fourth point I wanted to make.

So these are the rebuttals against the yen depreciation recommendation. May Iinvite your brief comment on each of them?

Professor Yoshino:
We are almost out of time, so I would very much appreciate it thank you.

Professor Stiglitz:
First on the question of China, obviously it is absolutely fascinating foreverybody everywhere in the world. I think one has to first appreciate thatChina is going through a process of restructuring its economy moving from acommunist system to a market economy, that while a big economy, is still a lessdeveloped economy. One of the attributes of being a less developed economy isthat it does not have a good safety net. It does not have unemploymentinsurance, not a very adequate pension programs. The only you might say thesocial program they have is full employment, or as strong employment as theycan. In my mind then, I view in part their exchange rate policy is a welfarepolicy. It is a peculiar way of putting it but it is a policy that has helpedthem create a lot of jobs that has maintained their social and politicalstability. Given the primacy that social and political stability is for theirwhole regime, I think they are gong to be very reluctant to engage insignificant devaluation.

It was interesting that even in the East Asia crisis, they did not devalue.Again, they were focusing on the issue of stability and I think that is acentral point in their mindset. Whether one would say is their exchange rateundervalued, should they appreciate, the answer might be yes in terms ofordinary market forces. But given their overall social and political positiontoday, I think it is going to be very difficult, although I think eventuallypressure will be brought to bear and something will happen.

I think the point that was made is that even an appreciation of China'scurrency would not solve Japan's problem and Professor Yoshino's statisticalanalysis showed that quite forcefully. It would make life easier but it wouldnot be a solution to the problem. One of the challenges that confronts the wholeworld is that right now China, you might say, has a comparative advantage in abroad range of manufacturing. That means there has been a change in the overalltheory of comparative advantage over the last 10 to 15 years. That means lots ofcountries will have to restructure their economies. This is a major globalevent. You might say, well China was suppressed for 150 years and now it is nolonger suppressed. However your interpretation, the fact is that it is a majorchange and it is causing problems. It is very interesting; it is causingproblems all over the world. Mexico after NAFTO was very excited about itsmanufacturing on the border, right now all those jobs that were created arebeing to China-not all of them but there is a real loss. So there is now inMexico a questioning of whether NAFTA was a good things:
"We got a few years of benefits but now we are facing huge problems ofsubsided American agriculture hurting the poorest of the people andmanufacturing being taken over by China." So this is a global problem.

One of the aspects of that, as one thinks of China having a comparativeadvantage in manufacturing is what are the things where China cannot compete?Basically those are non-traded goods and that included the service sector. Now Iknow that English is a problem but there is also a large service sector for theJapanese economy and there are inefficiencies in the service sector that isdirected at Japan so I think some of the structural issues that you emphasizedin your remarks, improving the productivity in the non-trading sector includingthe service sector, is part of what I would view an appropriate response to thenew global balance that represents China's entry into the global marketplace.

The final issue is all the concerns about the depreciation of the currency andwhat that will do. One of the things I try to emphasize is that again thesituation in Japan is different from say, that of Southeast Asian countries,where they were net debtors. When they depreciated the domestic currency values,their liabilities increased so depreciation has a very adverse balance sheeteffect. One of the advantages of Japan having run surpluses for so long is thatis a net creditor country and depreciation has just the opposite of increasingthe balance sheet. No obviously there may be particularly individual firms wherethat is not true, but overall in the aggregate it has a positive balance sheeteffect.

Now I think the issue you raised about the risk of capital flight is somethingthat one needs to bear in mind but my observation, my general view on this isthat crises are more often caused by failure to adjust in a timely way and that-maybeI am revealing the views of an American visiting Japan-but I always feel verypoor when I visit Japan because everything seems so expensive here. One way ofinterpreting that is I think your exchange rate in terms of standard purchasingparity is probably overvalued. If that is the case, in the long run there willbe an adjustment. And then the question is do you want to wait until it is a bigadjustment and you have a crisis when your debt GDP ratio is 180 percent or 250percent, or do you want to begin a more orderly adjustment now.

In my mind there are reasons to believe that the yen is overvalued and if thatis the case, this is a good time to begin because there are strong macroeconomiceffects. It will not solve the deflation problem on its own, you need otherinstruments as part of the macroeconomic strategy. As I said before, there isnot single panacea, no single instrument, one needs a number of instruments, butI think the depreciation for the currency would be part of the strategy thatwould end the deflation and help restart the economy.

Professor Yoshino:
Thank you very much for a very nice discussion. We want to keep on going buttime is already over, 10 minutes over. I would like to thank Professor Stiglitzas well as Special Advisor Kuroda. We were able to have active discussions.Thank you very much for your attendance.