Action, Reaction, or Over-Reaction?

Monday, December 1, 2008

New York City, NY

DAVID BRAUNSCHVIG:  Welcome to the Council on Foreign Relations.  This is a Roundtable on Business and Foreign Policy meeting, which we convene thanks to the generosity of the Bernard and Irene Schwartz Foundation.  This is an on-the-record meeting.  I would ask you to turn off your Iphones, Ipods, Itunes, Blackberries, and I'm probably leaving out a few brands.  And I want to suggest that, in this environment, to understand the complexity of what's going on is one thing we don't need is more quick slogans.

What we do need, however, is the perfect balance between – the  perfect combination of – excellence and balance.  We need, ideally, somebody who would have a multicultural sensitivity as a dual citizen -- for example, Canadian and American.   Somebody who would have majored in philosophy but also studied mathematics as a Rhodes Scholar.  Somebody who would have reached the pinnacle of academia with a Clark Medal and Nobel Prize, but equally at ease in boardrooms of companies large and small.  Somebody who had focused a lot on developed economies, but spent the last two years presiding a commission that focused on developing economies.  And, more importantly, somebody who can satisfy the emerging requirement for this particular roundtable:  Varsity hockey.  (Laughter.)  So, please join me to welcome Mike Spence.  (Applause.)

The title of this conversation is, "Action, Reaction, or Over-reaction ?", and most of you have heard of TARP, which is a pretty ugly-sounding acronym.  But it's not only ugly, it's really ambiguous, I find, in terms of its enunciation because…  What does it mean?  Does it mean Troubled-Asset-Relief Program or does it mean Troubled Asset-Relief-Program ? You know, this is something that we've been thinking of – and really, are our leaders acting thoughtfully?  Are they responding in an ad hoc manner?  How kindly, or not, will history judge our leaders?  What are the consequences, and the unintended consequences, of these policies?

The massive fiscal expenditures that have been already decided and will be in the incoming administration:   Will they be viewed as distant, good, bad, or terrible cousins of Sarbanes-Oxley?  Are we shooting right?  Are we mis-shooting?  Are we overshooting?  Are we undershooting?  Are we shooting ourselves in the foot?  These are our questions that we will address, but first things first:  Where are we today?  With the five-day streak of extraordinary gains in the stock market some people say that -- counted that -- (people love statistics in America in sports and stocks, everything ) that this was the largest five-day gain in 75 years.  Well, today it's a different story.

So what we hear -- I'm sure you hear that as well -- is a combination of “the end is near”, “this too shall pass”, or “gee, this is a great buying opportunity !”…  And sometimes you hear three in the same sentence, by the same person.  But let's talk later about prognosis and about therapy, and let's ask Dr. Spence his diagnostic.

A. MICHAEL SPENCE:  Well, I'll try to be very brief because I want to -- this room is full of people who individually and collectively know more than I do so I think the sooner we get to your participation in all this the better.  I believe that we were in a kind of slow burn, asset-deflation deleveraging process and the dynamics changed with extraordinary suddenness in September, and we had a credit lockup, so that's kind of part one.  A credit lockup that lasts destroys things more or less indiscriminately and I don't have any question that dealing with that by extremely unconventional methods was right.

I don't know how history will judge it but you can destroy an awful lot of businesses that are otherwise okay if they rely on going to, you know, short-term credit markets and discover all of a sudden they're closed.  That problem, by the way, has now spread to the rest of the world as of the last four weeks as capital, you know, rushed back to shore up balance sheets here.  With a little luck we've kind of dealt with that problem and hope that in the coming months that the private sector will take over again the supply of credit.

The more serious thing that we're in now is what you might call a sort of real economy asset-deflation sort of spiral in which balance sheets decline and that has an effect on consumption and then investment and employment and that affects profits and feeds back into the balance sheets, and we're going to, I think, try to stop this with a combination of coordinated fiscal stimulus globally -- messy but probably doable -- and something that is an attempt to slow down the asset side of the downward spiral.  My worry is that -- is we all knew that asset values were going to decline as part of the deleveraging process.  The problem is to prevent an overshoot that then feeds back into the real economy and keeps ricocheting.  And that, I think, is going to require unconventional methods including buying assets, you know, and then the question is feasibility -- which ones do you buy.

BRAUNSCHVIG:  As you talk about asset deflation, compared to what?  Twelve months ago were valuations realistic?  And if we're down from that level, is it catastrophic to be where we were three, four years ago?

SPENCE:  Well, I don't think anybody knows the answer to that, so let me just-- for the argument that you could overshoot -- what you want to do, is go down and level off at a level that looks like it's a reasonable guess as to the net present value risk adjustment of the cash flows, and going beyond that is an overshoot.  So you get up in the morning and look at DuPont, which looks still pretty healthy, and the dividend yield is 7.5 %  or 8 % and you think gee, it might be time for the value investors to arrive and yet they're still on the sidelines, for the most part.

BRAUNSCHVIG:  You are supportive of government intervention here, and yet my initial comments of the apparent haphazard approach…

SPENCE:  Mmm hmm.

BRAUNSCHVIG: … is that a good thing?  Is it normal that we should be doing a bit of trial and error here, trying to figure out what works, what doesn't?  Because it's pretty disconcerting for a lot of people, and maybe it's a communication issue or maybe it's a substance issue but some people would like to see a little more method…

SPENCE:  Mm hmm.

BRAUNSCHVIG: …in these policies, and you've been the advocate of a vision -- shared vision in these financial policy matters.  Is that something that's missing here?

SPENCE:  Yeah, it certainly is but the question is: Was it avoidable ?  I mean this:  You would have to characterize the process we're in as emergency responses, and I guess what happened is that sometime about the middle of September, the fire department was sort of working well and was about to run out of capacity putting out—fought—bigger and bigger and more and more frequent fires, and the chairman of the Federal Reserve went to the secretary of the Treasury and said we have to do something more systematic.  I guess I would say we sort of did, but not really.  The more systematic hasn't really come to pass, and there's at least two problems.  One of them is:  It probably would have been better with hindsight and it would have been better that we weren't in an interregnum between administrations.  It would've been better to put together a well-articulated sort of overreaction plan and communicate it.

One of the reasons investors are on the sidelines is that government's behavior is quite capable of wiping out asset values and it's quite unpredictable at the moment, and so that I guess was an unanticipated consequence of this sort of random let's respond to the latest emergency sort of approach.  It's not ideal. The question is: Could we have done better not having anticipated what we were getting into ?

BRAUNSCHVIG: Could we have done better and also, could we do better in the near future ?  In all likelihood, there will be a massive injection of government funds in the economy.

Not that long ago, a number of people -- including at your Growth and Development Commission -- were advocating responsible fiscal policies -- you know:   Deficits were supposed to be contained.  You know, People on both sides of the aisle were saying that here, and in Europe.  In Europe, the leadership in France and Germany are getting together to say, no, no, no, no --  we want to be able to spend more and go beyond the 3 percent, blah, blah, blah.  So everything that seemed critical a long time ago -- I'm talking about three months ago, okay -- seems to be completely forgotten.  So if we go on this spending binge, what's the limit of an acceptable budget deficit as a percentage of GDP?

SPENCE:  Well, I'm not sure I know the answer to that.  Look, I mean, you don't very often get a recession of this probable magnitude and depth driven by destroyed balance sheets, and so I think it's reasonable to worry first about preventing excessive damage in the short run.  And then -- and then more or less simultaneously, as Larry Summers said, you know, when -- when you -- when you go into the business of having a massive fiscal stimulus -- a countercyclical one -- then it has to be accompanied by a plan that runs out seven or eight years, that's believable, that puts you in the direction of restoring fiscal balance, and I think one without the other can be very highly destructive.

BRAUNSCHVIG:  We don't have that yet, and it would be nice if we could get it sooner rather than later.

The plan, so far, seems to be really--of the incoming administration--oriented towards creating public sector jobs.  Where is the private sector in all that?  At what point does  the private sector kick in and start really contributing to the economy ? Or are we talking… are we going to be locked into a post-Keynesian view of the world for a long time?

SPENCE:  Well, I mean, the hope is that the reason for focusing on public sector jobs is to focus on things that generate income that gets spent.  Then you hope the multipliers work -- the normal multipliers -- as opposed to, you know, something that gets saved and not spent.  This is a hard problem.  I mean, I think just designing fiscal stimulus that has the maximum impact is something a lot of smart people are going to have to spend some time on that --…

BRAUNSCHVIG:  Any suggestions?

SPENCE:  Well, I mean, I think spending on infrastructure, which is partly public sector but partly private -- it's implemented through private sector firms.  I think stabilizing the housing market is an essential item ingredient through interventions -- prevention of foreclosures, resetting mortgage terms, and so on.  I also think, you know, because of the rules that state and municipal governments operate under, that directly injecting resources, funds, revenues into the -- into the state and local level will help prevent what is otherwise going to be an accelerator -- meaning , a decline that runs through that side as well.  Jon Corzine said this–Corzine said this on TV the other day--I think that's right.  So those are three channels that one could look at.  There may be others.

BRAUNSCHVIG:  And bailing out entire sectors -- is that part of a plan as well and is there a rule or rule of thumb other than “oh, these people are screaming louder than others and therefore, you know, we have to listen to them” -- of why we should save one sector versus another?  It seems that picking winners and losers is not something governments have always been very good at and, in the automotive industry, you know, there are companies that have done quite well in Southern States, and others in the Detroit area are asking for help.  Is it a good idea to bail them out at the expense of their competitors?


BRAUNSCHVIG:  (Laughter.)

SPENCE:  I'll be very --

BRAUNSCHVIG:  …One-handed economist ?

SPENCE:  (Laughter.)  No.  So this is very uncomfortable for everybody, including economists.  Let me say just two things.  One, I believe there's about four sectors that are unique in the following sense:  that if they malfunction pretty much completely they bring down the whole economy -- and those are: finance, transportation, energy, and telecom.  And so there's lots of implications that if we had more time we could go into, but they have to be dealt with in a different way because their poor performance is in some sense a semi permanent public liability that needs to be kept under control.

Beyond that, I -- you know, there may be special circumstances, but I find the auto case extremely uncomfortable.  There may be a national security argument for intervention.  We know there's a lot of jobs that have to disappear under any scenario in the automobile industry, and so the question is: do you just sort of let the system deal with it ? Or do what apparently is happening or is maybe supposed to happen tomorrow which is the automobile companies are supposed to come (a la Chrysler many years before) with a credible plan and the government, if it believes it in theory anyway, is supposed to provide some interim financing at least for GM until they -- I don't think Ford is in imminent danger of running out of money.  But it is very uncomfortable in general.  I mean, if there weren't some kind of special argument for dealing with autos that goes beyond just as a big employer -- these three companies -- you know, what about retail?  I mean, retail is very labor intensive.  It employs a ton of people.  It's about to take a huge hit, and I haven't heard anybody suggesting we should bail them out.  So this is a very tough market.

BRAUNSCHVIG:  About 33 years ago my former partner, Felix Rohatyn, was pretty effective at exacting concessions from various constituencies when this city was going through problems.  Do you think the incoming administration will be better equipped, or worse, in that regard, at getting concessions, particularly from the unions in the Detroit issue?

SPENCE:  I think they will be reasonably tough and -- but whether they get concessions or not I think the more important thing is: They're tough and hold the line, and if they don't get the concessions then I think we ought to pay the price and let the -- and let the thing go.


SPENCE:  Through the process.  It'll probably be bankruptcy proceedings in the case of GM, and then we face the question who's going to buy a car from a company that may not produce cars in the future.

BRAUNSCHVIG:  Bankruptcy or a sale?  You know, the U.K. automotive industry -- you have a lot of jobs there.  You just don't have a nameplate that's owned by British interests.  But –people don’t seem to be complaining.

SPENCE: I think that’s fair.

BRAUNSCHVIG : We talk about policies but should we talk about institutions as well?  Are our institutions well equipped to deal with this set of issues ? -- you know, that some people are saying that it's impossible for our current government agencies to be really insulated from political pressure, and that we should have a Reconstruction Finance Corporation-equivalent like what was created by Hoover and then expanded by Roosevelt.  Is that a good idea or a bad idea to create new institutions, at this point?

SPENCE:  I think by and large, it's a bad idea.  I mean, you know, we are going to create some new institutions, including some new regulatory ones, but at least I'm in favor of sort of thinking long and hard about it and not doing it in a rush.


SPENCE:  I don't think the political pressures are necessarily destructive of having a positive, you know, useful, even unconventional response to the current circumstances.  I mean, it takes time.  I mean, it took two tries to pass TARP.  There were a lot of things added to it.  But the fundamental insight that part of it needed to be focused on homeowners, housing, transparency, balance sheets, and toxic assets-- what are now called toxic assets,  I think was a good instinct.

BRAUNSCHVIG:  Is this something we should think of in an international context?  We're here at the Council on Foreign Relations -- it's probably time for us to talk about that.  There was a G-20 meeting recently.  One of the very few conclusions -- tangible conclusions -- at that meeting was: We must do the best we can to lower trade barriers.  And so, two days later, Russia increased its tariffs on imported automobiles; and three days later, India increased by 5 percent its tariffs on steel and other products.  So, can we trust these international gatherings to do anything but produce words that have no effect?  Again, same question as previously -- should -- as you know, the French president has been very vocal in favor of creating new Bretton Woods and new institutions to deal with it.  Are there sectors where it makes more sense than others to have new institutions to regulate this problem on an international basis?

SPENCE:  Well, let me take the G-20.  I -- they didn't -- the Doha Round is on life support I guess and I have no idea whether it'll make it back.  What the G-20 said, was that they kind of committed themselves to not using protectionism as a way of dealing with the crisis because it's a “beggar thy  neighbor” approach, and the examples you've cited were not consistent with that statement.

You know, there are other examples of this that are a little more forgivable.  Many of you know that in the commodity price spike, the poor people in the developing world are basically starving because they spend 50%, 60 % of their income on food and if, you know, the price of rice doubles or triples that amounts -- is tantamount to a huge cut in income.  And what happened in most of the developing world or a good chunk of it was they didn't have any policy instruments -- you know, no conditional transfer mechanisms to get resources to the -- for the poor.  So they did things that viewed from a distance were highly dysfunctional.  Thailand and India both put bans -- export restrictions -- on rice.  This is exactly what you don't want to have happen in the global, you know, grains market, if you want a big supply response.  On the other hand, they were a little short of policy instruments.  I think this will change over time but -- but so -- but nevertheless there is this kind of dysfunctional behavior.

In the financial world, I think it's absolutely essential that we have a good long rethink and probably some rebuilding of international institutions for two reasons -- one, we're interdependent in a, in an obvious, severe way in the quality of oversight regulation of financial stability, at least in the large countries; and secondly, there's some global issues, like the imbalances we had before that can't be dealt with except on a kind of coordinated multinational basis, if we're going to avoid a mess.  So on both grounds, I think -- no, I had -- nobody I guess that I know had any sympathy for the 90-day or 100-day time horizon for constructing these institutions.  That seems a little precipitous given the complexity of the challenge.

See , I think we're going to have to sit down and understand how we built that much systemic risk into a system, sort of semi invisible because if we don't understand that, then we don't have the kind of fundamental underpinning for deciding, you know, how we're going to measure instability, systemic risk, and then build regulatory structures around them.

BRAUNSCHVIG: But if we have -- if we want countries to express solidarity with regards to rescues, bailouts, regulation -- if you push that logic, would you see a political consensus in this country to have the U.S. taxpayer bail out foreign companies?

SPENCE:  No, not -- only in a system that was designed, you know, to accomplish some appropriate level of burden sharing, which we simply do not have.  Let me just make a point I made this morning.  Unlike the United States, there are a number of countries in the world that have multinational global financial institutions whose balance sheets are big in relationship to the country's GDP, and if they get into big trouble it's not clear that, you know, you can solve that problem, you know, on a single country basis, and that's a real risk.  Now, the European institutions were on balance much more highly levered than ours and they haven't fully kind of hit the wall yet, and a number of them are big enough -- I mean, the extreme case, of course, is Switzerland with UBS and Credit Suisse.  I forgot what number but somebody totaled up the balance sheets and compared it to the GDP of Switzerland and it didn't look promising.  So -- so big global institutions like this are going to need something other than single countries dealing -- and we're certainly not going to take care of the whole lot -- the United States.

BRAUNSCHVIG:  Another thing these institutions need, is a different way of managing themselves and governing themselves.  You’re not anymore, but when you were a director of Bank of America how did that pan out ? And -- and what -- how would you envision creating a little more responsibility, accountability for people who are, by law, fiduciaries: The directors?

SPENCE:  I don't -- In my personal opinion, and I may be quite wrong about this, is that the problem is not a sense of accountability.  It's the -- it's finding effective ways to discharge that responsibility.  I mean, if you -- Bob Rubin said, and there's an important element of truth in this -- that if you weren't on the trading floor you actually didn't know what was going on.

BRAUNSCHVIG:  That's a problem, if you're a director, isn’t it?

SPENCE:  Yeah, it is, but I'm trying to distinguish between intent on the one hand and how you get at, you know, responding to the responsibilities you clearly acquired by virtue of being a director or a senior member of management, and these problems that are linked to the one I mentioned before:  I think that we need, sort of, academics and practitioners with long years of experience to sit down and really think about the question, how does a system of this complexity, you know, build or not build systemic risk.  It gets really, I think -- to me it's the central question.  There's lots of components of it -- incentive structures, disclosure requirements, regulated partially, fragmented -- all that is relevant and they're all pieces of the puzzle.  So I don't mean to dismiss them but I think the fundamental question is can we -- can we do what Gordon Brown said:  Build a global early warning system and that, if the answer to that is yes,  it means that we can look at a system with a bunch of experts and see if it's starting to become unstable.

BRAUNSCHVIG:  Would you chair it?

SPENCE:  I think there are better people than I to chair such a thing .

BRAUNSCHVIG:  Six months ago, we -- we had meetings here, and I can guarantee you that the following themes were not the only thing that was addressed but a good part:  Some people were complaining about global imbalances, right?  That this country doesn't save enough, other countries save too much, and we consume, and they export from Asia, and we can't export to Asia, and this is all terrible.  So you had report after report of the IMF and World Bank on global imbalances identifying this problem as a key one.  You had energy costs going through the roof, commodity costs, food costs, the specter of rampant inflation very close by.  A lot of these issues have been checked, so:  Are we dealing with a silver lining, in that this is perhaps a correction that's too brutal, too violent, unexpected, not time but ultimately, will have positive effects?

SPENCE:  Yeah.  I mean, Ned (Edmund Phelps) knows perfectly well.   Among economists -- we didn't even agree that there were global imbalances and the -- or the question was how long and is there a natural way to get out of this if you think the global imbalances were defined primarily by the low savings rate in the United States and a collection of high savings rates mostly in Asia plus the Gulf countries, who have to run surpluses for obvious reasons.  So the answer is yes, I mean, to the extent that one thinks we were out of balance.  I think it's a little deeper than just that pattern.

This is a particularly ugly way to get out of it and furthermore, I mean, if we -- if China does what they say they're going to do, which is drive the savings rate down, it reached the ridiculous level of 55 percent of GDP and over 10 percent above the investment rate which is 45 percent of GDP -- I mean, this has never happened before.  And then -- by the way, there's no strategic reason whatsoever for a developing country to save more than it invests.  That's foregone consumption for people who are relatively poor with the proceeds lent to the United States government at 3 percent.  It is just -- there's no argument at all for it, and why they -- and China didn't do it until the last three years and then they made a set of policy mistakes -- with the benefit of hindsight – let it run up.

Anyway, they're going to bring it down and it's a big economy -- $3.3 or $3.4 trillion -- and some others will bring it down too by using the domestic economy to replace the foregone exports that, you know, disappeared on our side.  So we'll come out in better balance and if that pattern sticks there's no reason why we can't go back to global growth rates in the 4 to 5 % range that is what we had before.  Just a different mix of growth drivers -- and probably a healthier one.

BRAUNSCHVIG:  Well, thank you.  On that note, I would like to open it to you.  Please raise your hand and when I identify you please introduce yourself.  And if you have trouble making a difference between a statement and a question, I will help you.  (Laughter.)

QUESTIONER:  Farooq Kathwari, Ethan Allen -- Where do you think we will be six months from now, and 12 months from now?

SPENCE:  I'm hoping that six months from now that the capital markets will start to stabilize, meaning the volatility will go and people will have a clear -- clearly defined plan from the federal government and a set of similar plans around the world – but I think we will be in a quite deep and fairly extended recession because of the deleveraging requiring time to run through.  And this is a personal opinion -- and I'm going to defer to experts on the kind of macroeconomic things.  I don't think we'll be out of it in 12 months.

BRAUNSCHVIG:  Mr. Gutfreund.

QUESTIONER:  Hi.  When you emphasized the importance of financial structures around the world, I was thinking that it's very difficult for any management to understand the complexities of all the subdivisions.  You've talked about it from the point of view of a director.  In terms of manager of these businesses, either the businesses are too big or the structure is not managing.

I think it's the same thing I've said, but it's a question of -- I don't see -- Citibank is a good example to me where that company, with all the arms and legs, nobody could manage that, in my judgment, not even the best team.  I don't know how you reconcile that with progress.  It's a question.  (Laughter.)

BRAUNSCHVIG:  Thank you.

SPENCE:  Yeah.  So, I mean, all of these things, John, seem to me to revolve around the same thing.  The sensible response is, in the short run, when one doesn't understand something, is not do it on the assumption there may be hidden risks in it.  But that, of course, isn't the long-run answer, because it stifles the highly legitimate part of risk spreading.

I mean, securitization is a wonderful way to spread risk and lower the cost of capital, but it got out of control.  But I don't have a good answer for you.  I mean, some of it is going to be probably putting restrictions on systems that have a cost in terms of efficiency.

Take certain derivatives that are sort of tailor-made.  They can't be traded, can't be netted out, and therefore there’s a huge transparency problem trying to track through the counterparties in the system.  We may have to give up on that, even though there's a reason why those contracts were tailor-made for  particular circumstances.

I'm not an expert at these things, but I think the frame of mind we're going to have to go in with is, on the one hand, we want innovation:  We want, you know, new structures that accomplish financial things.  But on the other hand, we have an interest in a system that's understandable enough so that the leaders of it, managements primarily, with some support from boards, can actually understand it as well.  It is a huge challenge.

QUESTIONER:  Peter Fisher from BlackRock.

SPENCE:  Hi, Peter.

QUESTIONER:  Let me ask you about the optimal sequence of recovery.  You've addressed the downward spiral, balance sheet, income statement, spilling over to income and employment.  What's the optimal sequence to come back?  Until the last few months, Chairman Bernanke was pretty precise that we wouldn't recover until we saw the housing sector recover.  So I just --  there could be a sub-optimal path.  But what, in your view, will be the optimal path for the sequence of recovery?

SPENCE:  Hmm. I'm not sure I can guess at the exact sequence, but right now I think we need to have both the fiscal stimulus and an asset purchase program that's feasible, that includes -- and this is partly political as well as economic -- includes mortgages and the securities that were based on it.

It probably goes beyond that.  And I realize -- I mean, one of the reasons TARP got sidetracked was it didn't work fast enough, and it fortunately was written in such a way that direct infusions of capital into financial institution balance sheets was possible.  And it may have been impractical.  I mean, it was a structured investment vehicle, so I guess the actual assets are held in a trust.  And if you want to get your hands on them, which is what you'd have to do to deliver on the intent of the legislation, you'd have to buy all the tranches.  And that just sounds like it's too complicated.

So, I mean, some hard work in the next month or so on what assets we could feasibly buy that had the effect of delivering on some relief for homeowners who have mortgages they shouldn't have undertaken, reducing foreclosures, and some effect on both the balance sheets and the transparency issue, that's the goal.  And the original TARP  may have not been the right answer.

I mean, I thought it was pretty beguiling.  I mean, Warren Buffett and Bill Gross said, you know, "If you lend me money at 3 percent and give me enough time, I'll go and figure out what these assets are worth and I'll buy them for 6 percent less and make a nice return for all of you, and proceed."

I guess the fly in the ointment in that formula is time.  So we may have to buy things that are simpler to buy, even though it's not the ideal.  I mean, I'm not proposing this  -- you know, in the currency crisis, some of you will remember that Hong Kong bought one-third of the equity market in Hong Kong because they were so worried about this, you know, back and forth between the balance sheets, on the one hand, and the real economy on the other.

QUESTIONER:  Hi.  Dan Drezner.

Can I push you on the question about the macro imbalances?  You suggest that the silver lining to the crisis is that it might sort itself out.  I'm a little more pessimistic about that.  It's nice that China wants to lower the 55 percent savings rate.  There's no evidence to date they've actually been successful in the past at trying to lower their savings rate.

And more importantly, the U.S. has to presumably raise its savings rate, but at the same time that it's doing that, it's also going to be running a massive fiscal stimulus.  Furthermore, the trade imbalance is going to be made more problematic by the fact the dollar has actually increased in value.

So I'm sort of curious.  How do you see, in the long run, that sorting itself out?

SPENCE:  So those are hard things to sort out.  First of all, on the Asian side, I think China -- it's a hard problem, because China has to do several things.  They can use fiscal measures to lower the savings rate in the short run, just as we're proposing to do.  And they have an enormous amount of resources to do that, you know: A fiscal surplus, rising revenues, enough reserves to sink a battleship, and so on.

China did not run big surpluses until 2005, when they started to rocket up.  But most people don't actually know that.  They were running up reserves, but that's different.  Reserves were run up because they were holding the currency against the dollar and they had a huge net inflow of private capital, which was either going to drive the value of the RMB up or get reversed.  They reversed it.  I mean, there's a lot of controversy about that, but that's what happened.

Social insurance systems are not very well built, and so the private saving is very high.  But it's only about 40 % of the total savings.  The other is corporate, and that will be driven by profit opportunities, and presumably will decline a little bit.  And government is a huge fraction of it because of all that infrastructure they built.

So the jury's out on how fast they can get it down and whether they can build enough social insurance so people aren't -- even poor people aren't saving at ridiculously high rates.

On our side, I think what you just said is right.  I mean, it's going to take a while, but what looks like is going to happen is the American consumer is going to save a bit more.  So that piece to the puzzle looks like it'll fall into place, you know, unfortunately, in the way we're now observing.

BRAUNSCHVIG: Michael Schrage.

QUESTIONER:  Michael Schrage. Thanks.

Your own work focuses on signaling, in your Nobel work.  And in your answers in the op-ed that you distributed, there is a question here of to what extent are these government actions more important in terms of the actual substance of what they do, versus the signals they are sending to the markets.

Is there any sort of way, over the next six to 12 months, you'd like to see a handover between the signaling value of the actions that you recommend, and the actual impact they have on market behavior?  Or are the two indistinguishable in your own mind?

SPENCE:  No, they're not.  I mean, they may be indistinguishable empirically, but, I mean, there's a real difference.  There's a coordination problem.  I think most people agree that if everybody suddenly magically decided that now is a good time to start investing, at least on the asset side it would be -- and the government is the entity that gets to try to solve the coordination problem.

The other thing that makes it similar to (this is a little esoteric; I apologize for the other people in the room) it makes it similar to signaling-like models -- that is, models in which people's beliefs about the world are endogenous -- is that -- which is a characteristic signaling models have -- is that if you think things are going to go down, you'll take actions that cause them to continue to go down.  That's why it's a coordination problem.

Nobody would call what we're in an “equilibrium”.  But whatever you want to call it, it's very likely there's multiple versions of it -- thought of as multiple places to bottom out.  And what the government is trying to do, in addition to directly changing, you know, aggregate demand, on the one side, is change the expectations so that the behavior changes in a self-confirming way.

QUESTIONER:  (Michael Schrage)  Is that more of a “rational expectations” thing, or is it really a behavior, true behavioral economics ?

SPENCE:  No, that part is more like rational expectations.  I think the behavioral economics is going to make a contribution in understanding the way people think about risk and react to it.  The instinctive -- I know this isn't a good description of sophisticated risk modelers, but basically, at the risk of oversimplifying, I think what happened is that people started out with risk models that assumed more or less exogeneity of risk, stationarity, meaning it's not moving, and low correlations, which is the underpinning of both insurance and diversification models.  And all three of those things turned out not to be true.  It was endogenous, non-stationary and became increasingly correlated over time.

And figuring out how that happened, I think it's going to take behavioral economics to understand how people react to data in relation to the model they start out with.  I mean, the insidious thing about this process, if  you start out with the wrong model that no data got thrown out, you know, to disconfirm it of any significance until the thing broke.  So you had to start out with a different model.  I mean, and it's figuring out what that means with a lot of smart people, both academics and practitioners, that I think is a big challenge.

BRAUNSCHVIG:  If I could follow up on that question, would you consider back-testing those hypotheses?

SPENCE:  Yes.  But remember, I mean, as Nassim Taleb said, even if you think the thing is stationary, you know, in low-probability events, you need an enormous amount of historical data.  And if the system is non-stationary, then you kind of don't know what you're doing, I mean.

BRAUNSCHVIG:  Nick Gardiner.

QUESTIONER:  Nick Gardiner, Gardiner International.

It seems that trust has evaporated.  What does it take to re-establish it?

SPENCE:  Hmm. Well, it has evaporated to a considerable extent, but not completely.  I think people still trust the United States government in the sense of, you know, being willing to invest in Treasury bills.  We could lose it by irresponsible behavior.  But trust in the financial system is one of the things that I think is severely damaged on a global basis.  Trust in the kind of global system is at least damaged.

My prediction is that there'll be two responses to that.  One is defensive behavior by investors, by countries, by just a lot more attention to risk mitigation on the assumption the system is just more volatile than we thought.  In the meantime, there'll be a rebuilding process that's going to involve building institutions that people come to believe in.

The hard part of  this is that effective multinational institutions are going to not only have to build trust, but they're going to have to have the authority to act and the resources to do it.  And we don't have any evidence yet that nation-states are willing to give up that much authority.

QUESTIONER:  Hi. It’s David Einhorn.

I've got a few questions related.  What makes you think that the government is bigger than the problem in that there's enough resources of the government to buy up all the assets that need propping up?  And related to that, what happens if it turns out that the problem is bigger than the government, if we go about buying up all these assets?  And in that case, why aren't we better off saving our national treasure rather than propping up asset prices and using it to actually help individual people rather than propping up financial institutions?

SPENCE:  Because -- I think the short answer to that is, because you need the financial system to perform essential functions like credit.  And if it stops doing that, then you can't go out and help people, you know, in a way that's effective.  I mean, that's what the credit lockup was all about.  That's why people, you know, gritted their teeth and started, you know, rebuilding the balance sheets in the financial sector.

The American public clearly didn't want to do it.  That's why TARP failed on round one.  You know, people thought, "Well, let's just leave them alone.  You know, they caused this problem.  We'll just leave them over there.  And then we'll help these people." (Chuckles.) And unfortunately, the system just doesn't work that way.  Now, that doesn't mean you can't do both, but it means you can't just sort of skip the financial system, I think.

BRAUNSCHVIG:  The gentleman right here.

QUESTIONER:  John Train.

At the end of the remediation of this problem, how inflationary do you expect the remediation will have been over the long term?

SPENCE:  Again, Ned (Edmund Phelps) probably has a view on this.  My view is that right now it's not inflationary, because basically what's being done is we're replacing -- well, fixing balance sheets but replacing, you know, credit that's been withdrawn.  And you can see the same thing in the developing countries.  I mean, take a country with large reserves.  Basically all they're doing is taking money that's rushing out and replacing it by money that's going back in, and then a few other mechanisms to make sure it's accessible as credit.

Now, if you do it too long, and so they're busy doing it, you know, while the private credit is coming back, then it'll become inflationary immediately.  So it's a matter of judgment and balance.  That, by the way, is the hope that we won't overextend the government's resources, is that if you do this for a while, eventually private capital will start flowing back in and take over again.

QUESTIONER:  Could I ask you about one of the companies that the government has stepped in to save, the first one, or one of the first ones, and that was AIG, and what you think of sort of the various incarnations this has gone through, and especially the fact that that's the only company the government has taken ownership of, basically, 80 %?  Do you think that ownership will actually last over the long term?

SPENCE:  No.  (Chuckles.)  I might say a different thing if we were in France, but -- (scattered laughter) -- no, I -- maybe it's hopefulness, but, I mean, I think the government will engage in a process of selling the assets it ends up acquiring over a sort of reasonable period of time, so as not to disrupt markets.

Now, there's a separate question.  You know, the first takeover, we already owned it, kind of, which is the GSEs, right? And -- as the week before.  And I don't know what's going to become of them.  I mean, that's sort of an open question.

BRAUNSCHVIG:  Gentleman right here.

QUESTIONER:  Thank you.  Mark Angelson, R.R. Donnelley and MidOcean Partners.

Mr. Spence, there has been some discussion of a government-guaranteed automobile warranty of 50,000 miles or five years, and the implications that that would have for people all along the supply chain, without making the political decision to favor Michigan over the states in the South.  Do you have a view on whether that's a viable proposition, at least?

SPENCE:  I hadn't heard of that one.  (Scattered laughter.) So I'll give you my first reaction.  It sounds a little nuts to me. (Chuckles.)

QUESTIONER:  What's your second reaction to it, please?  (Laughter.)

SPENCE:  Well, I mean, the reason it sounds nuts is, you know, it's a lot more complicated than just guarantees.  I mean, there's, you know,  commitment.  This is trust again, right?  I mean, when you buy a car, you're buying, you know, a company that's implicitly said they're going to be around long enough.  And so guarantee is one thing that might vanish, but it's only one.  So that doesn't sound like a -- it sounds like a cute, incomplete solution, I guess is what I would say.

QUESTIONER:   Bryan Lawrence.

You just mentioned Hong Kong as an example of where they bought a third of the stock market and then backed away.  I think other people have talked about Sweden as a successful example of government intervention.  And the gentleman in front of me asked the question about deflation versus inflation.  I mean, I think a lot of that is whether or not the government knows when to stop.

So what gives you confidence that the government will know when to stop?  And what lessons, maybe, can you draw from prior experiences by Government as to what our government might be doing to make it more easy for it to stop?

SPENCE:  You know, one of the most interesting fields of economics right now is political economy; that is, the interaction of sort of political forces and economic forces.  You know, it's associated with young guys like -- well young compared to me -- like Daron Acemoglu and others.  And Daron received the John Bates Clark medal.  And, you know, they'd be better positioned to answer this.

I don't really know, you know, that there's a good answer to the question other than ultimately the wisdom of the American people and our elected officials.  So I don't think government is sort of some other thing.  You know, in the end, it's kind of us and the people we elect.  And by and large, although we all makes lots of mistakes, I don't see any reason to believe that they'll kind of run off more or less indefinitely and do something silly.

Now, some of these things require very sensitive judgments as to timing.  This inflation/deflation issue is one of them.  And it's very easy to see that we could start to build in an ugly inflationary environment if we just kind of went too far.  And there's lots of examples of it.

You know, Argentina, which is, I think, about to have another crisis, you know, had a lovely growth experience after the 2002 crisis, and then they got near potential output.  And instead of allowing the central bank to put the brakes on, they started using price controls.  And pretty soon they had, you know, controls on exports, because people like eating beef, and the farmers, you know, blocking all the roads and the demonstration.  And they most recently seized all of the private pension fund assets.  This is -- so I cite this as an example of there's no guarantees, you know, about sort of sensible government behavior.

In the developing world, one of the most common, quote, "mistakes" (policy mistakes), is to develop -- you know, is,  to discover a successful formula -- a hard thing to do -- that drives very high growth and stick with it too long.  You know, usually it's sort of labor-intensive something, manufacturing, that eventually, you know, ceases to be a growth driver because you're pricing yourself out of the market.  And there's a very, very strong tendency to keep doing it too long.  So I'm trying to give you some data to support your skepticism.

BRAUNSCHVIG:  We have time for a couple of quick questions and quick answers.

QUESTIONER:  Ray Debbane.  The Invus Group.

This crisis came from excess leverage, as you described.  In a way, it was a credit market bubble.  Unlike previous bubbles, however, they were not driven by the greed of the people buying this time, because people who were buying Triple A paper were only trying to make 50 basis points higher interest in low interest-rate environment.

What made it possible was, to some extent, the failure of the credit agencies – Moody’ss, S&P, etc -- that's how it was with thoseTriple A ratings.  Is this a failure -- should we consider this a failure of the market because they're private entities, or a failure of regulation because government basically mandates what pension plans should be buying in terms of Triple A, etc, etc?

What's the lesson for the future?

(Cross talk.)

BRAUNSCHVIG: That seems to be binary – one or the other. (Chuckles.)

SPENCE:  Good. We'll have to choose.  No, it was clearly part of the system that -- it was private, and part of the system that underestimated the systemic risk.  I think it's pretty easy to understand how that happened, because of the complexities that we talked about before, but it was on the private-sector side.

BRAUNSCHVIG: David Malpass

QUESTIONER:  I'm David Malpass.

A follow-on to that, then.  Was the low interest rates of 2003, '04 and '05 part of that systemic problem?  But I also wanted to ask you about -- right now Treasury is borrowing at the short end of the curve, for relatively short maturity, and buying the long-term assets.  We'd be more stable if it borrowed longer, given the low interest rates available way out in the curve.

SPENCE:  I think they're thinking of doing that, but I'm not sure.

QUESTIONER:  Can we trade on that, or -- (laughter) -- why did you say that?  I haven't seen any mention of it?

SPENCE:  No.  I mean, I don't remember off the top of my head, but maybe there was somebody like you suggesting it, and they're not even thinking about it at all.  But it sounds pretty sensible.

QUESTIONER:  (Off mike.)

SPENCE:  Yeah, right.  (Laughs.)  No, just very briefly, yeah, I think that the low-interest-rate environment, which was in response to the Internet crash and then 9/11, surely, you know,  without being quantitative about it, contributed to an environment which was conducive to using high levels of leverage.

BRAUNSCHVIG:  Last question here.

QUESTIONER:  Steve Treadway.

What is so complex about allowing financial institutions to be leveraged 40 to 1 at the same time that they have balance sheets which essentially have substantial assets which are borrowed short and invested long?  Is that really a complexity question, or is that a fiduciary question, a regulatory issue?

SPENCE:  I don't think that’s where the sort of complexity lay -- I mean, you'll forgive me for guessing, but, I mean, the complexity lay in the complexity of the securities that appeared to make risk go down, whereas in fact it hid it, with the benefit of hindsight.  And because it was hidden, the leverage went up.  It appeared justified.

The SEC in 2004 allowed the leverage (inaudible) to go up to 30 to 40 to 1.

SPENCE:  That's correct.

QUESTIONER:  Wasn’t that a massive change in risk?  Isn’t that a simple issue?

SPENCE:  Not if the risk is relatively low.  I mean, if you actually have low risk, you lower the cost of capital by using leverage.  Furthermore, if leverage is causing an asset bubble, then you can't use the normal ratios to measure the appropriateness of it, because both its numerator and the denominator are kind of out of whack.  I'm just trying to tell you that I think this is actually a little harder than you make it seem.  But surely that -- the removal of that constraint, with the benefit of hindsight, contributed to this.

BRAUNSCHVIG:  Well, this is harder than it seems.  But thanks go to you, Michael Spence:  You’ve made it seem a bit simpler for us, this evening.  (Applause.)
















Top Stories on CFR


The U.S. health agency regulates the country’s foods and drugs, among other products. The COVID-19 pandemic and controversy over vaccines and treatments have highlighted the FDA’s role.


Evaluating the legacy of September 11, 2001.


The recent coup in Guinea is part of a resurgence of "coup culture" in West Africa.