[TRANSCRIPT PREPARED FROM A TAPE RECORDING.]

Present at the Press Briefing were:

Mr. David Goldsbrough, Acting Director, IEO

Mr. Marcelo Selowsky, Assistant Director, IEO

Mrs. Isabelle Mateos y Lago, Senior Economist, IEO

P R O C E E D I N G S

MR. GOLDSBROUGH: Good morning. Let me make a few opening remarks. I'd first like to say a couple of things about the IEO, who we are and what our procedures are, and then I'll give a brief summary of what we see as the main messages of the Argentina report, and then I'll throw it open for questions.

First, on role and procedures of the IEO, I want to mention a couple of things because there has already been some speculation in the media that has perhaps misunderstood our role.

We operate entirely independently of IMF management. We conduct a small number of evaluations every year, three to four evaluations a year, on topics where we can generate lessons for the IMF's operations. And our work program is essentially chosen by the Director of IEO. It's not dictated by anybody outside. We obviously consult with people, including the IMF's Board, but the final choice is that of the IEO.

And by mandate, by the terms of reference of the IEO, we do not involve ourselves in ongoing operations. This means we do not involve ourselves in current programs. And that's important in the Argentina case. That explains the cutoff for the Argentina report, you know, where we assess the Fund's role from the beginning of convertibility throughout the 1990s and during the crisis period, up to the collapse of convertibility in the first few days of 2002. We do not look beyond that date because it would involve commenting on what could still be regarded as ongoing operations.

In terms of how we handle evaluation reports, including this one, we essentially have very similar procedures to the U.S. General Accounting Office. When the IEO was first set up, we talked to many evaluation units to find the best way of doing things to try and guarantee our independence. And the GAO procedures seemed to be the ones that would be good to follow.

So, when we have finished a report, we send it formally to IMF management for comments. But when we send it to IMF management for formal comments, that version is set in stone. We do not change the report as a result of any comments we receive. And, in fact, when it's sent to IMF management for comments, it is also copied for information to what is called the Evaluation Committee, which is a subcommittee of the IMF's Executive Board. And, in fact, through them it is copied to all Executive Directors. So Executive Directors on the Board get the report at the same time as management gets it for formal comments.

And then what happens is that management-sometimes it is management and staff separately-make comments. Their comments along with any IEO response is all then part of the package that is discussed by the Board of the IMF. And then that entire package along with the summing-up of the Board discussion is what is made available, what is published.

We don't change the report at all except for pure factual errors. You know, we got the name of the capital city wrong or there were some numbers that were wrong, et cetera, and then we actually issue an errata sheet to the Executive Board. And there have been no substantive changes to the Argentina report. A few typos, that was it.

That procedure, which is different than many other evaluation offices of the IFIs, was deliberately set up so that we don't get into the business of negotiating the language of the report. Even if we agree with management/staff comments, you know, even if in retrospect we think, well, they were right and what the report says was incorrect, we would just reflect that in our response to their comments. We wouldn't change the report that is published.

And I should also mention that in those procedures, there's also a provision if the authorities of the country mainly concerned--in this case, Argentina--wants to make comments, then they are published along with the package as well. And in this case, Minister Lavagna did make comments, although I should stress that we're evaluating the IMF's role in Argentina. We're not evaluating Argentina per se.

This Argentina report in particular was cleared and approved by Montek Ahluwalia, who was the Director of IEO until June 30^th-in fact, he cleared and approved it on his last couple of days in office before he resigned to take up a position in the new Indian cabinet. And, you know, that affected the timing. There has been some speculation in the media about the timing of the report. In fact, he was very anxious that he be the one to finalize and clear the report, which is why it happened right at the end of June.

Okay. So much for procedures. Let me turn to the main messages of our evaluation report.

In terms of the assessment of the fundamental causes of the crisis, it is our judgment that it was essentially a combination of the failure of Argentinean policymakers throughout the 1990s-and during the crisis period of 2000-2001-to take the necessary corrective action to make sure that domestic policies were compatible with the choice of the exchange rate regime.

Now, of course, this isn't a particularly new conclusion, but the fundamental problem of policymaking was the inconsistency between fiscal policy, particularly the size of the public debt and the fact that most of it was borrowed externally; the incompatibility between that and the choice of exchange rate regime-the hard peg, convertibility regime that essentially constrained other macroeconomic management choices. And the incompatibility between a rising public debt and increasing borrowing and this exchange rate regime left Argentina vulnerable, and then when a long series of adverse shocks hit, beginning '99 through 2000, that essentially caused a slow-moving crisis that eventually led to the collapse of the regime.

As I said, that's not particularly a new conclusion. Rather, the main objective of our evaluation was, with that background, to assess what was the role of the IMF.

And bearing in mind that this is a case undoubtedly where the fundamental policy choices were owned by the Argentinean authorities, the different governments at the time what did we learn about the role of the IMF? The evaluation identifies two major weaknesses, two sets of areas in which the IMF's role in Argentina should have been different.

First is in the context of the surveillance and program relationship of the Fund with Argentina during the 1990s. And I'll say more about that in a moment. And the second is during the crisis management period itself from late 2000 onwards.

Weaknesses in both of these areas in surveillance, longer-term program relationship, and in-crisis management, also reflect problems with the decision-making processes in the IMF.

First, on the 1990s, it's clear and we analyze in some depth in the report that surveillance, pre-crisis surveillance, significantly underestimated the vulnerabilities that Argentina faced, particularly under-estimated the vulnerabilities arising from a growing public debt, and in several ways. First, because of a focus each year on the flow levels of the fiscal deficits, it didn't pay enough attention to what was happening to the path of public debt, because there were many other things going on that increased the public debt. The social security reform, pension reform was done in a manner that imposed additional longer-term fiscal burdens. There were all sorts of judicial decisions that ended up being financed by issuing additional debt. Similarly, the treatment of privatization receipts was done in a manner that didn't sufficiently recognize the capital nature of some of those receipts in the early days. And so, you know, that's a problem of analysis, if you will. There wasn't sufficient attention to those issues.

And, secondly, just the level at which the public debt for a country like Argentina becomes a problem-the thresholds at which warning signals should have been given-should have been set much lower. We know now from quite a bit of work, including work done in the Fund, on debt levels that countries that are heavily reliant upon external borrowing, the danger thresholds can be at much lower levels. So surveillance really didn't do enough to catch those problems.

Still with the 1990s, the program relationship between the IMF and Argentina, particularly in the second half of the 1990s, didn't succeed in tackling the most fundamental problems that were underlying Argentina's growing vulnerability. And by "fundamental problems," I'm referring to what you might call the fiscal structural problems, dealing with the problems of the relations between the federal government and the provincial governments, dealing with the social security--you know, the fiscal consequences of pension reform, dealing with fundamental problems of weak tax administration, which have gone back for decades in Argentina.

These problems were recognized. It wasn't in this a case a problem that the Fund, or the authorities, for that matter, didn't recognize that these issues need to be dealt with. But as an actual matter, the Fund-supported programs were inadequate to deal with these problems--inadequate in two senses: in their initial design, they didn't do enough to address them; and there were, in fact, many slippages in the programs. So, actual fiscal deficits turned out to be systematically higher than targeted deficits.

And we have in the report a few charts that show what happened with each program, and the extent of the slippages in Argentina's case were much larger than an average of what's happened in Fund programs.

And, of course, all of that meant that vulnerability was being built up; the debt was being built up. And it meant that when crises hit, it was much more difficult for Argentina to carry out countercyclical fiscal policy because as soon as there were pressures, questions were being raised about the long-term sustainability of the debt.

So that's one set of weaknesses, both, as I said, in surveillance and the program relationship during the 1990s.

Turning to the crisis management issues, the conclusion the report comes to is that the initial decision to support Argentina with a large package, the so-called blindaje, in January 2001, was understandable. In thinking about all of these decisions, one has to think about them in a probabilistic sense, that there's always going to be risks that they will or will not work, and then essentially any ex post evaluation has to ask: Was it a reasonable decision at the time? Was there a reasonable chance that the strategy would work given the costs of the alternative?

And using that sort of criteria, we come to the conclusion that the initial decision to try this approach, to try to support Argentina in January 2001 with a large package, was understandable. But there wasn't sufficient consideration at that time or later about the "what if" question. You know, what if the strategy doesn't succeed? What will be the contingency? What will be the alternative? And there wasn't sufficient thinking about the criteria under which you decide has the strategy failed, which we refer to in the report as stop-loss guidelines, something like that. You know, when do you decide that, all right, this strategy was worth trying, but it's not working and we need to reconsider?

So our main message about the weaknesses of the Fund's role during this period was essentially not with the initial decision in January 2001. As I said, judged in a probabilistic sense, it could be justified. But it was with the lack of sufficient contingency planning, and because of that the decision to continue supporting a strategy that under most reasonable judgments one could have concluded, even with information at the time, was not working. Because of these problems and an understandable concern on the part of the IMF that whatever alternative was adopted was going to involve very painful consequences for Argentina. Clearly the decision makers were very aware that, you know, any alternative would be very costly.

Because of that, they essentially got locked into a strategy that wasn't working. And it would have been better--again, hindsight is easy, but we think even with the information available at the time-it would have been better to have had an earlier change of strategy.

As I said at the beginning, these diagnoses of the problems in the Fund's role, both in surveillance, prolonged program involvement, and the crisis management, reflect certain underlying problems in the Fund's decision making processes. I have already mentioned the lack of sufficient contingency planning, the lack of clear criteria for judging when a strategy--you know, a change in strategy was needed. A third area is the role of the Executive Board, and for various reasons, the Executive Board did not fully perform its oversight function.

Now, let me elaborate what we mean by that because I think, again, in some of the media reports that we've seen already about this issue, there's perhaps some misunderstanding.

What the report says is that both during the surveillance period and during the crisis management period, what you would have hoped to have seen is that the Executive Board was examining in-depth on the basis of full information, candid assessments from the staff, some of these key sensitive policy decisions.

During surveillance, for instance, is the exchange rate regime consistent with Argentina's other policies? And if not, what needs to be done?

During crisis management, well, all right, we've made a decision to support Argentina with a large package, essentially supporting continuation of the present policy framework. But what are the alternatives? What are the costs of the alternatives? Under what circumstances might we need to change such strategies?

There wasn't sufficient discussion in the IMF's Board of those key strategic decisions, for various reasons, including sometimes lack of full information. You know, the information given to the Board by the staff was sometimes essentially supporting the preferred chosen strategy rather than laying out alternatives, and an enormous concern for confidentiality on sensitive matters.

It's very clear that one thing that was motivating both management and also the Board in some sense was the recognition that these were highly sensitive issues, and if they were openly debated and there were a leak, you know, they would have caused self-fulfilling crises. And it was those considerations that underlay in part the fact that in the end the Board didn't fully perform its oversight functions.

So those are the main lessons, the main messages. Let me just say a few words about what recommendations we make.

We have in the report, six recommendations, but let me just focus on four of them that I think are particularly important.

The first recommendation--and this goes back to the surveillance period--is that there's a need for strengthened, more candid surveillance of exchange rate regime choices by countries. And this is a message that goes well beyond Argentina, of course. I may add that, in fact, it's a message that the Fund has already recognized. There's just been recently--every two years there's an internal review of the Fund's surveillance operations, so-called biennial review of surveillance. And that review comes to the same conclusion, looking at more general cross-section evidence, that the Fund doesn't raise the awkward questions about whether a particular exchange rate regime is consistent with the other policy framework countries are pursuing. And the reasons are much the same as those I've mentioned, you know, the sensitivity of the issues. So there needs to be much more candid surveillance of exchange rate regimes. That's the first message.

The second message is to avoid prolonged program involvement if policies are not being undertaken to deal with the fundamental problems underlying a country's vulnerability. And in a sense, that message goes back to the message that the IEO gave in its very first evaluation. The very first evaluation we did after we were set up was of the prolonged use of IMF resources. At one point we'd contemplated actually doing a case study on Argentina as part of that prolonged use evaluation, but we decided that it would overwhelm the rest of the studies, so we deferred it to later. But the message is very clear from the Argentina case, too, that many programs that are not dealing with the most fundamental structural problems underlying vulnerability--and in this case it's fiscal structural problems--are in the end not very productive. And so, you know, the Fund should perhaps not have had so many programs with Argentina during the 1990s.

The third message, turning to the third recommendation, is that the Fund needs better contingency planning, and the report talks about the need for clearer stop-loss guidelines. Now, by that we don't mean some rigid rule that, you know, if reserves hit a certain level, then all bets are off. It can't be a rigid rule, with no scope for discretion. But there needs to be clearer markers as to, well, what are the criteria for success or failure of a particular strategy to force the system to consider what the alternatives are under such circumstances.

We're under no illusions that in the case of Argentina that would have been difficult because the then-government was firmly committed to the current policy strategy and obviously did not want to contemplate alternatives. But it's the Fund's job, you know, when it's helping in these crisis management situations, to think about those alternatives and to engage the authorities in such a discussion.

And then, finally, the report has a set of recommendations to strengthen the Board's decision making processes, which are essentially directed at making the Board the locus of all key decisions. Clearly it already is in a governance sense, but to strengthen its role, strengthen, you know, the candor of the information it gets and make sure that all of these key decisions that I referred to earlier are essentially debated in the Board.

Some of that might require additional arrangements for confidentiality of particularly sensitive decisions. I mean, there is already a procedure for dealing with so-called side letters, when there's particularly sensitive questions, and it may be that that needs to be elaborated further.

Thanks for your attention. Let me finish with one last thought.

I'm sure many of you are going to say, Well, what does all of this mean for the IMF's relations with Argentina now? As I said at the beginning, by our mandate we don't involve ourselves in ongoing operations. And the way we see the lessons from this report, they're general lessons. In fact, you could say in many senses they're lessons that are much more applicable to other members than Argentina, you know, Argentina's current situation. They're how to deal with surveillance in vulnerable cases, how to handle crisis management when there are pressures and uncertainties as to whether it's a liquidity or a solvency question.

Because we stop at the first few days in 2002, we are not dealing with, well, how do you handle the consequences of debt default? How do you deal with the aftermath of that? That's perhaps a subject for another evaluation, but it's not the subject of this evaluation. So in that sense, there are perhaps fewer lessons in this report for the current situation in Argentina than there are for, you know, the Fund surveillance role and crisis management role generally. But we think the lessons are very important because, although some people have said, well, the situation in Argentina was unique--yes, you know, every situation is unique and the size of the problems in Argentina were much larger than in many other cases. But these sensitive and difficult decisions that the Fund faced in this case, it will face again in other situations. And it is important that the decision making processes be as good as they can be for dealing with what will inevitably be very difficult situations.

Thanks. Let me stop there and throw it open to questions.

QUESTIONER: Heather Scott with Market New International. Two quick questions.

First of all, who did you talk to when you were making this evaluation? For example, did you speak to the former Director of the Western Hemisphere Department when having discussions on why they supported the exchange peg for so long in the IMF program? And, also, you mentioned it would have been better for the IMF to have had an earlier change of strategy. Is there any recommendations on what that strategy should have been specifically?

Thank you.

MR. GOLDSBROUGH: Since you ask a question about who did we talk to, let me ask Mrs. Mateos y Lago, who was part of the core team doing the evaluation, who she talked to.

MRS. MATEOS Y LAGO: Well, we tried to interview most everyone within the Fund who was involved in Argentina during that period, so that's a lot of people, both current and former Fund staff members. And we also interviewed about 200 people outside of the Fund who are listed in the report. We didn't list Fund officials, but all the others are listed, and so, yes, we did speak with Claudio Loser, since you mentioned his name.

Should I also start answering on the alternative strategy?

MR. GOLDSBROUGH: Sure.

MRS. MATEOS Y LAGO: Well, we don't make a prescriptive judgment on exactly what the alternative strategy should have been because there's a large amount of speculation involved in that. We simply don't know what would have happened. But what we do say is that by, you know, the spring of 2001 it became fairly clear that the convertibility regime was simply no longer sustainable, and that it would have been preferable to cut the losses at this point and take an active stance towards getting out of this regime.

Now, exactly what that should have involved and, in particular, towards what kind of exchange rate regime to go, we don't make a judgment on that. In fact, you know, the IMF staff spent hours and hours and hours trying to figure out whether Argentina would be better off under a dollarization or under a float or under dollarization after a devaluation. They, in fact, never came to a definitive conclusion on that.

In our view, what would have mattered is to go with whatever scenario the authorities would have been prepared to sign up to. But one of the key problems is that this discussion between the Fund and the authorities never really took place--in fact, never took place at all, not in the spring and not later either.

So, really, the key point that we're making is that the Fund should have been working on alternatives and should have discussed with the authorities and with IMF shareholders on that rather than keep buying time and funding a system that was obviously no longer sustainable.

MR. GOLDSBROUGH: Let me just supplement that. In terms of the information that we drew upon as Isabelle said, the team talked to just about everybody involved. Also, one of the IEO's comparative advantages is access to all internal IMF documents, both those that go to the Board and those that circulate just within the staff. That's true for all evaluations, and it was true for this one. So we looked at all of those.

The only thing that we do not have access to is if there are--and, again, this is standard procedure; it's not just for the Argentina evaluation--is if there are exchanges between, you know, the Managing Director or part of the management team and a Minister or something like that and, you know, he writes a note of that meeting. There's a zone of privacy there that we don't have automatic access to unless he chooses to share it with us.

However, if memos are then written to the staff on the content of such communications, that we do have access to. So essentially we have very wide access.

Isabelle's right that there are many possible permutations of alternative strategies. In the report, there is a box that lays out one possible counterfactual, you know, what might an alternative strategy have looked like.

It's very clear, however, that the Fund faced--Fund management faced--very difficult decisions as it was doing this, and in their mind was--I mean, they were thinking about the political economy aspects, not just the economic aspects, and they were very aware of the problem of how do you get from Plan A to Plan B in a political sense. Is the Fund's actions in trying to encourage a move from Plan A to Plan B going to be such that it affects decision making or the political situation in Argentina in a way that, you know, you're going to end up with even more disorderly policies?

So there was a lot of thinking about, well, how do you exit from a strategy that's not working in a way that makes for, you know, reasonable decisions in formulating an alternative. And so the report lays out one possible alternative. But we're very aware that, of course, that's easy to do. You know, it's a bit like the famous joke about, you know, economists always assume they have a can opener. So you can always assume that, well, the politics would allow you to adopt such a set of policies, but the big question is: Would the politics have allowed it? And, you know, that we can't answer.

QUESTIONER: Mark Dragen from Bloomberg News: One thing that you haven't touched on, which, of course, has become famous from Mike Mussa, is the August program. He called it, I think, the worst decision in Fund history. Have you come to a definitive stance on that?

And, secondly, just this whole idea of kind of cutting Argentina off earlier, is it your assessment that if that had happened, the crisis in Argentina that followed in 2002 would not have been as bad? I mean, what's--what exactly was lost by the IMF supporting Argentina throughout 2001?

MRS. MATEOS Y LAGO: On the August package, we're absolutely clear in the report that we think it was a mistake. And, in fact, we're also saying that the May disbursement, you know, the Third Review under the program, was also a mistake in our view because the situation was unsustainable, clearly unsustainable.

Now, that doesn't mean that we think the Fund should have cut Argentina off, as you say. What we say is that the Fund should have supported a different strategy at that point.

Now, of course, if the authorities had not wanted to change strategy, perhaps the result would have been cut off. But this is not what we're suggesting. We're suggesting that these $9 billion essentially that were disbursed between the spring and September could have possibly or hopefully been put to more productive use if they had been used in support of, say, a move to Plan B, a devaluation, or something like that, could have been used to limit the overshooting of the currency depreciation, and in the meantime would have avoided further deterioration in banks' balance sheets, would have avoided another six months of a really sharp recession.

Now, it would still have been very costly, it would still have been an ugly recession, but we think less so than ended up being the case. Also because if the Fund had managed somehow to remain engaged with Argentina, then perhaps the decisions that were made in the very first few days of 2002 that were not helpful to the change of regime, you know, things like the asymmetric pesification and all that, might have been avoided.

Now, you will have noted I used a lot of "perhaps" and "might". We simply don't know. But there's a number of ways in which a better outcome could have been achieved. It would have been an ugly crisis anyway, but perhaps not quite as bad if the Fund had supported a change of strategy earlier.

MR. GOLDSBROUGH: Marcello, you wanted to--

MR. SELOWSKY: Let me elaborate on this and in the process answer the earlier question of the lady.

An alternative would have been not to cut off financing but, for example, support the alternative strategy of an exit from the peg, restructuring of the external debt, further fiscal adjustment, and use the external financing to dampen any extra costs like the overshooting of the exchange rate, as Isabelle mentioned, but also to soften some of the hit that the banking sector would have suffered exiting the peg. So that instead of using external financing to maintain the peg, that financing could have dampened some of the costs of exiting.

MR. GOLDSBROUGH: And, in fact, it's that sort of strategy that we talk about. There's this box in the report that asks, well, what might an alternative strategy have looked like. But I do stress my point about the politics. In that context, the question is: If the Fund had pushed for such an alternative approach, could it have succeeded in getting--you know, encouraging a decision making process in Argentina that would have allowed for a more orderly but still very costly exit? The report says it should have done that, but, of course, it's impossible to predict what would have happened on the political economy front in such a counterfactual.

Yes?

QUESTIONER: Yes, Pablo Bachelet, Reuters: I understand that you can't really go into the current program between the IMF and Argentina, but in his statement commenting on this report, Lavagna clearly sort of draws a lot of parallels and conclusions that obviously favor the current Argentine position. He says there is a lot of blame to go around and that Argentina seems to be paying most of the costs of those mistakes, and he wants some sharing of that, of those costs. So, first of all, I'd like you to comment on that.

Second, on the issue of reforms, which seems to be a key issue today, even now in Argentina, it was an issue in your report, and apparently the Fund wasn't sort of strong enough in demanding some of those reforms. Well, they seem to be pretty strong now, and Lavagna is complaining about that.

So I was just wondering, I mean, in your--what does this report mean in the reform sense for the program? Should the IMF continue to be strong?

MR. GOLDSBROUGH: Well, as I said, we're not going to comment on ongoing programs. We have no more information than anybody else, including all of you, to allow us to do so. So, we just haven't assessed that.

On this issue of sharing the blame for the costs, et cetera, there's a fundamental difference between choices a government makes and an organization like the IMF supporting those choices.

All I can repeat is the message of the evaluation report that the governments at the time during this process clearly owned the fundamental policies that were taken, you know, the convertibility regime, the basic fiscal choices that were taken, you know, fiscal federalism type issues, the choices on the pension reform, the choices on various modifications to the tax system. These were all taken by the governments at the time. They certainly weren't forced upon the governments by the IMF.

In fact, there's a joke--it isn't our joke; it's been said before-that at some times, particularly during the crisis management period, the policies were so owned by the Argentinean authorities that they didn't even tell the Fund what they were doing. And that's true at certain points, particularly during the Cavallo period. There were many choices that were made and just announced unilaterally.

So although, as we've said, there were significant weaknesses in the Fund's role and lessons that need to be learned, I don't think you can look at that in a context that says it implies consequences for burden-sharing? You know, governments decide and governments choose, and that's very different than the role of any other institution. That's the first point.

And the second point is that you have to recognize the nature of the business that the IMF is in, particularly during crisis management periods. As I said, these were always probabilistic judgments. The IMF is trying to support a country that's got a problem, that's got a problem with maintenance of the policy regime that it's firmly committed to. And everybody knows the cost of a shift is going to be huge. And, you know, the Fund's trying to--the Fund tried to sort of help the country with its strongly preferred strategy. The report's judgment, as we said, is that at some point a change of strategy was necessary. But that doesn't mean to say that you want the IMF to be necessarily always supporting sure things. It's possible you could, of course, think of an entirely different architecture for the international financial system in which there isn't any sort of crisis manager, crisis lender. But as long as the choice is made, you know, to have the present architecture, you can't have a system where you extrapolate from one case in which, you know, mistakes were undoubtedly made. And even if they hadn't been made, not all attempts are going to be successful, to then say, well, somehow that implies very different things about burden sharing. I think that's just drawing too much from the Argentina case, although as we do say, we think that the Argentina case does have important lessons for the Fund about how it goes about decision making.

Yes?

QUESTIONER: Harry Dunphy from AP: From your opening remarks, if I understand them correctly, you seem to be saying that this was sort of a collective failure of both management and the Executive Board, and that one of the constraints on the Executive Board was confidentiality, if this got out there would really be a problem.

But were you able to identify anyone in management or any member of the Executive Board who was saying we've got to stop, we've got to change, or was this just sort of an institutional snowball that nobody could halt?

MRS. MATEOS Y LAGO: Well, in August, as you know--that information has been in the public domain for a while--two Executive Directors abstained, so you could at least guess from that that they were beginning to have serious doubts about the validity of the strategy.

Prior to that, there were every now and then a couple of Directors asking questions, and, in fact, from December 2000 onwards, at every meeting of the Board, whether formal or informal, you had a couple of individuals saying, well, we ought to be thinking about an alternative strategy, you know, a Plan B.

But at no point was there really any discussion of when do we decide to go to Plan B, and by the way, what is Plan B? Because every time the issue came up, well, management's answer was to say, well, you know, we're working on that but we shouldn't be discussing it because this is really too sensitive, and if this were to go out, it would be a disaster.

So that's more or less what happened, and Directors took that answer at face value and were comforted by it sufficiently to accept to go ahead.

MR. GOLDSBROUGH: Just let me add to that in terms of the internal--you know, the decision making process in the staff. You'll see from the report which draws upon these internal documents that I referred to. There clearly was a lot of debate internally about, you know, is this the right thing to do, what to do, et cetera, and there were different views expressed. So by no means was the institution monolithic in these things.

Of course, at some point, you have to decide on the strategy, and then the question is: How do you make sure that the institution as a whole, and particularly the Board, is weighing sufficiently alternatives to that strategy, but in a way that isn't, you know, rocking the boat and undermining the chances of success of the preferred strategy? And, that's at the core of how to handle the decision making process, that we don't think there was enough exploring-at the Board level--of alternatives.

QUESTIONER: Antonio, AFP News Agency: You wrote that the crisis resulted from the failure of Argentine policymakers and that the January 2001 decision to maintain the support of the IMF was justified, but not thereafter. Does that mean that the government of De La Rua and Mr. Cavallo were responsible for what happened and that earlier Argentinean leaders had nothing to do with it?

MR. GOLDSBROUGH: No, I don't think you could draw that conclusion. As we said, the seeds of the vulnerability that eventually led to the crisis were laid during the 1990s. So, there's plenty of responsibility to go around in that sense.

Going back to this diagnosis of the nature of the crisis, vulnerabilities were created and increased throughout the 1990s, essentially because of a mismatch between the choice of exchange rate regime and other policies that built up over time. And there is a lot of analysis in the paper of that issue.

These vulnerabilities came to a head because Argentina was hit by a whole series of adverse shocks at a particular time, you know, the events in Brazil which changed the Brazilian exchange rate, the strengthening of the U.S. dollar, changes, conditions in global capital markets, et cetera.

Now, that series of adverse shocks hit at a particular time and affected the timing of the crisis. You know, if it hadn't been then, it would possibly have been some other different time. But the vulnerability was built up much earlier, and as we say, it was essentially--well, there are many aspects to it, but the, major aspect was the failure to address these fundamental fiscal structural problems.

QUESTIONER: Rich Miller, Business Week: I wonder if I could just ask a couple questions. And I apologize, I haven't had a chance to read the report. But is there any discussion about the political--you mentioned politics in Argentina. Is there any discussion about sort of political pressures or potential political pressures that might have come from the major--major members of the IMF Board onto the staff and management?

Two, you mentioned that the IMF perhaps was more tolerant of slippages in the Argentine case than in other cases. Any reason why that may have been?

Three, you seem to be suggesting that less transparency in some cases is a good thing. You're talking about--I think you alluded to side letters, which I gather--I'm not quite that familiar with them, but I gather those would not be published, perhaps. That obviously goes a little bit against the grain of what has been the sort of common wisdom here at the Fund about more transparency being a good thing. I wonder if you could comment on that.

MR. GOLDSBROUGH: Okay. Let me speak to a couple of things, and then perhaps I'll ask Isabelle to supplement as she wishes.

On the political pressures from major shareholders, the report does actually talk about some of those issues. The real problem with the decision making and the role of the Board was the one that I referred to earlier. It's the nature of information flows, candid thinking about alternative strategies, et cetera. And in that sense, it isn't a question of some shareholders versus others. That's the first point.

The second point, while--as I said at the beginning about the nature of the information we have, we don't have access to any sort of conversations that management had with individual Ministers. So while recognizing that we don't have 100 percent information, it is not our judgment that, you know, this is--you know, the Argentina case is a story that somehow one or a group of shareholders forced a particular solution on the Fund that was wrong, et cetera, et cetera. That's not what the report says. Those involved at a senior level in the decision making at the time made very clear to the evaluation team--and this is said in the report--that they did not feel constrained to take the decisions that they thought were the right decisions by any such external pressures.

So although, of course, more generally in the Fund there are all of these issues about voice and the role of G-7, et cetera, that's not what we're talking about here, not what the report focuses on. But there are issues about making the Executive Board effectively the locus of decision making, and the report does talk a lot about that.

On the question of tolerance of slippages, why more in the Argentina case? The evaluation report does say one part of this was that for a while the Fund's program arrangements with Argentina were precautionary, in other words, there was a presumption that they would not draw upon them.

Now, in a procedural sense, there's not supposed to be any distinction between the strength of the program and the way it's implemented--you know, the way it's monitored, et cetera, between a precautionary program and regular programs where countries are expected to draw. But it is the conclusion of the evaluation team that in this particular case, the fact that it was precautionary and that, therefore, Fund exposure was planned to go down, et cetera, did mean that there was a more accommodating approach to slippages.

Now, whether you can generalize that to all precautionary arrangements, that's perhaps a bit of a stretch from just one case, but it does seem to have been a factor in this case.

On the transparency issue, take, for instance, the exchange rate and, you know, a general conclusion which everybody, including the Board, shares that there needs to be much more candid surveillance of exchange rates. I think there'd be several elements to that. One is that you want to have better analysis, better tools. I'm including forward-looking tools, building into debt sustainability analyses, et cetera. And we do make a number of recommendations in that area. And all of that is in a sense making available to everybody, including the markets, better information, and that certainly should be public and transparent. And the IMF has already made considerable moves in this direction already.

But at some point, when the Fund as an institution wants to have a dialogue with a country and then think itself about is there a fundamental problem here in a particular country and what needs to be done about it, I mean, it's very hard to imagine that you can have that debate openly. One of two things would happen. One, nobody cares what the IMF says, in which case it has no impact. Or they do care and it risks causing a self-fulfilling crisis.

So there has to be--and, you know, this is something that the Board itself is going to be thinking more about as they go forward--a fine line there between transparency and this need for confidentiality.

MRS. MATEOS Y LAGO: I'd like to add two things on this issue of political pressures. First, one thing that is clear, especially when you look back, is that there was quite a bit of political pressure on the IMF's decision making process as a whole. I mean, throughout the summer you had declarations by various heads of state or government, you know, being broadly supportive of Argentina, not necessarily saying the IMF should disburse X billion dollars by this date, but, you know, broad expressions of support that inevitably put some pressure on the IMF, be it its shareholders or management or staff, it doesn't matter.

Secondly, you asked specifically about pressures on the staff. I think there's no question that IMF staff felt pressured. It said so in the internal postmortem on this Argentine crisis that was published not very long ago, that the Board discussed at the end of last year, you know, the internal review--I mean "lessons from the crisis." There's a sentence there that said that, you know, they felt pressured. So--but that's just a perception that is out there. Then you have to be clear what you are talking about. Some staff members think that any decision by the Board is, in essence, political or more political than what they would have it. So that, again, doesn't necessarily mean that something else took place.

And, thirdly, it's clear that there is a bit of a gap between the internal assessment that, you know, staff were writing and sharing with management and what they wrote in the staff reports that went to the Board and were published. We document that in the report--and you could argue that this reflects some pressure.

But, again, on the fundamental thing of whether management's arms were twisted by some shareholders or others, we got a very clear message from members of the then-management team that it wasn't the case.

QUESTIONER: Maria __, TPA: I am under the impression--and I would like to know if I'm correct--that the main problems were in the implementation of the program, because what the report says is there wasn't enough surveillance. I'm not talking about the 2001 crisis management, but before; that if the authorities of Argentina had implemented the program and the compromises that they assumed in the stand-by agreement regarding the fiscal deficit and the structural reforms, then maybe it would have worked.

So is that correct? Was it a problem of implementation and not of in the beginning on the design of the theory of this program?

MRS. MATEOS Y LAGO: I would say in part you're right. I mean, in part it was a problem of implementation in the sense that there were lots of waivers. But, sometimes, and especially looking backwards, it's hard to disentangle the two, because throughout there was a notion that programs should be designed in such a way that you shouldn't ask from the authorities things that they wouldn't be able to deliver. So in the end, it all becomes endogenous. You know, in the negotiations, the Fund tended to--at least in this particular case, the Fund tended to internalize the political constraints under which the authorities were operating. And so program design was probably less ambitious and less demanding than it should have been, in particular with lots of the structural reforms being in conditionality in the form of structural benchmarks, which are not really binding. So if you do it, fine; if you don't do it, you don't even need to ask for a waiver. So that's one thing.

But even so, there were a number of occasions where the formal conditions were not met, and in those cases the authorities generally got waivers. So it was really both, but I'm not sure that the distinction is all that meaningful.

MR. GOLDSBROUGH: Just to supplement that a little. We don't want to create the impression that we're saying that, well, always just there needed to be more fiscal tightening and that was the only thing, the only solution. What we're saying is that there was a need for much deeper reform of the fiscal--you know, the structural components of fiscal arrangements in Argentina, both fiscal federalism, tax administration, dealing with the pension reform, et cetera, all guided by a recognition that there was a vulnerability because of the rising public debt. And for various reasons, most of them associated with political economy factors reasons within Argentina, those reforms never really got implemented. Early on there was significant progress, but it sort of tapered off and they never really got implemented.

And, although the IMF was certainly aware of what needed to be done--it made some analysis mistakes in underestimating the risks of the public debt, but, in terms of the structural fiscal reforms, it certainly knew what needed to be done. But for the reasons that Isabelle referred to, essentially the constraints got endogenized in the design of the programs, many of these fundamental issues were not solved. That's one message.

The second thing that happened in Argentina, which happens in other cases, too, is that there ended up being an asymmetry in the implementation. You know, when growth was strong, fiscal performance didn't get better. Often it didn't even achieve the initial targets. And when growth was weak, targets were adjusted. Targeted deficits were adjusted and waivers given, et cetera.

And all of that, the repeated experience of that over a decade left--Argentina with a much larger public debt, more vulnerable, which meant that when this large series of exogenous shocks hit, several things happened. First, it was in a deep recession, anyway, because of the operation of the convertibility regime, you know, which tends by its nature to mean that you have larger cycles in such circumstances. And it couldn't carry out countercyclical fiscal policy because immediately there were concerns about the sustainability of debt. So there were pressures to almost carry out procyclical fiscal policy--not because anybody had a simplistic idea that, oh, well, more fiscal adjustment is always good. It was just the nature of the constraints that Argentina faced at that point, which left it with nothing but unpalatable choices.

QUESTIONER: Yes, I just want to know if you met Mr. Cavallo. Yes?

MRS. MATEOS Y LAGO: Yes, we did.

QUESTIONER: Did he explain why he took some decisions without informing the IMF?

MRS. MATEOS Y LAGO: Well, we promised confidentiality to everyone we interviewed, and that applies to Mr. Cavallo as well. But I think on this particular point, it's very clear that he thought it was his program, he owned it very much. The perception was that the authorities should be in charge, which in reality the Fund accepts as well. Then there's really just a question of how you implement that sort of broad principle.

MR. GOLDSBROUGH: Okay. If I can make one last little remark, just to respond to one issue that has come up in some of the Argentinean press. I didn't say it in my initial remarks because it would have sounded a bit defensive, but let me make it at the end.

There were references to, well, we spent $6 million on this evaluation, da, da, da, what did we get for it, et cetera. My response is I wish we had $6 million. The entire budget of the IEO for all projects-and, as I said, we do three to four a year in addition to publishing an annual report and doing other outreach activities. The entire budget is about $3.3 million. And that information is publicly available. You know, it's in our annual report.

So those estimates are off by an order of magnitude, but as I said, I wish they were right.

[Laughter.]

Questioner: How many people worked on the report?

MRS. MATEOS Y LAGO: The contributors are all listed in the report, six, seven people, but then if you want, you know, full-time equivalent, I don't know exactly to how many bodies that amounts.

MR. GOLDSBROUGH: Yes, the core team was three to four people. Thank you.

[Whereupon, the press briefing was concluded.]