Global imbalances continue to place the stability of the global economy at risk. The International Monetary Fund’s forecasts anticipate essentially no reduction in existing imbalances in the next five years, assuming the continuance of current policies. And independent observers have suggested that, if anything, the IMF may be overly optimistic about the prospects. A shock that causes those imbalances to unravel quickly could lead to sharp drops in the currencies of countries that depend most heavily on foreign finance for their current-account deficits (“countries that depend most heavily on foreign finance” being code for the US). That in turn could have major repercussions, both real and financial. As the IMF puts it, the global economic recovery could be “resting on hollow legs.”
Admittedly, not a few of us have warned before about the risks posed by global imbalances and pointed to savings-investment imbalances in the US and China as their source. Some of those warnings were issued as early as 2004. That these early warnings were — how to put it politely? — premature does not mean that they were off target. They were derailed by the global financial crisis, which directed attention elsewhere. Evidence that global financial markets were seizing up had the effect, ironic in the circumstances, of inducing additional flows into the dollar, that currency being a traditional safe haven in times of financial turmoil and the US treasury market being the most liquid in the world. But simply because these warnings were early and rendered moot for a time by other events does not make them wrong.
The G20, rightly, has continued to make global imbalances and their correction, along with international monetary reform and measures to foster growth in the poorest countries, a focus of its deliberations. At their February 2011 meeting in Paris, G20 finance ministers agreed to a set of three types of indicators on which the sustainability of national economic policies would be assessed: public debts and deficits, private savings and debts, and current-account balances. The composition of this set was a compromise between US concern that global imbalances reflected Chinese policies toward the current account and the Chinese contention that they were a function of excessive private-sector debt and public-sector deficits in the US. That said, the compromise was not undesirable, since the three categories of indicators effectively cover the imbalances waterfront.
At their subsequent summit in April in Washington, DC, G20 finance ministers then agreed that the IMF would take four approaches to identifying levels of these indicators that were problematic:
- A structural approach based on economic models (presumably including the global economic model maintained by the IMF and used for the forecasting in the World Economic Outlook);
- A statistical approach based on country-specific historical experience and trends;
- A statistical approach that compares national positions with those of other countries at comparable stages of economic development; and
- A statistical approach that gauges the sustainability of national approaches relative to the experience of other G20 countries.
When conducting these exercises, the G20 agreed that data for 1990 through 2004 will be used. Countries identified as having “persistently large imbalances” according to at least two of these approaches will then be subjected to a detailed assessment both by the IMF and the countries themselves, where the IMF will use its own standardised data, while individual countries undertaking peer reviews will be permitted to use their own national data. This assessment is intended to identify the root causes of their imbalances and the impediments to their correction. Larger countries whose policies are likely to have more powerful impacts on their neighbours, meaning those that account for at least 5% of the G20’s collective GDP, will be held to more demanding standards and subjected to closer scrutiny.
This language is an interesting combination of the overly specific and overly vague that is typical of international agreements. Specifying four approaches to assessing whether or not imbalances are a problem, some of them in considerable detail, is exceptional for a finance ministers’ declaration. This is clearly designed to limit the discretion that can be exercised by the IMF when it undertakes the exercise. It is designed to similarly limit the ability of the countries that are the subjects of the exercise to dispute the IMF’s first-round findings. From the point of view of those concerned about imbalances, both constraints are a good thing.
But the G20 was overly prescriptive in specifying 1990-2004 as the only period relevant for gauging what sustainable is. To be sure, the choice is not entirely without logic; international capital flows were limited prior to 1990 (the 1980s being the period of the developing country debt crisis); and the period after 2004 is widely viewed as one of clearly unsustainable imbalances. But who is to say that prior historical experience — that of countries like the US, for example, that have long-run trade deficits and had open capital markets — is uniformly irrelevant and uninformative? Indeed, some of us are precisely in the business of attempting to draw out the implications of that prior experience.
On the vague end of the rhetorical scale, words like “excessive” and “unsustainable,” which have clear economic and political meanings, are not used in conjunction with imbalances. “Persistently large,” the language that the G20 does use, can mean very different things to different people. For those who feel little urgency about correcting prevailing imbalances, it can reinforce their sanguine view.
Moreover, where a considerable amount of detail is provided on the four metrics that will be used to determine which countries are singled out in the first round, there is little meaningful detail about the criteria that the IMF will use when identifying the root causes of imbalances and recommending corrective action. When determining not just whether but also why some imbalances are so “persistently large,” will it use its multi-country model, for example, or rely on statistical approaches utilising time-series or cross-section data? Will the data used in its analysis be limited to specific years and its cross section comparisons to particular countries, as in the first stage? What weight will the IMF place on these different analytical approaches when recommending corrective action?
How the IMF will proceed at this second stage may have been left deliberately vague because identifying the causes of imbalances and the policy measures needed to correct them is even more difficult analytically and contentious politically than simply determining when an imbalance is large. There was no way, pragmatically speaking, that G20 countries could agree on the particulars.
But, pragmatics aside, there is an analytical inconsistency between the approach taken at the two stages. It is not really possible to identify, in the first stage, when an imbalance is a problem without having a view as to why it arose. The latter requires either agreement on both causes and consequences — that is to say, there must be agreement on an explicit analytical model of the determinants of imbalances — or else it requires a willingness to delegate the decision about how to make that determination to an entity like the IMF. And if governments are willing to hang their hats on a particular model or key set of relationships in the first stage, it is not clear why they should be reluctant about doing so in the second stage.
A more cynical take would be that countries were willing to agree on explicit procedures that left the IMF with little wiggle room at the first stage because there was no chance that the IMF and the G20 would force them to do anything at the conclusion of the second stage. The IMF has long offered polite suggestions for policy adjustments on the part of its members. But, understandably reluctant to bite the hands that feed it, it rarely uses strong language where its large shareholders are concerned. Even if it were more forceful, there still would be nothing to compel compliance, in particular by large countries that issue debt in their own currencies and therefore have no need to borrow from the IMF.
To see this, one need only recall the last time that the IMF engaged in this kind of exercise. A Multilateral Consultation Initiative was established in 2006 to bring together a handful of countries, under IMF aegis, for consultations on issues where their policies matter jointly rather than separately. The 2006 consultation focused on global imbalances and involved the US, the Eurozone, Japan, China, and Saudi Arabia. While it was from all reports an interesting exercise, at its conclusion the countries involved all simply returned to business as usual. Why should the current exercise be any different?
The answer is that in the short run it shouldn’t. A list of countries where the G20’s warning lights are flashing red will presumably be issued in the fall. Sometime after that, the IMF and the members will issue findings on the causes of the imbalances cited on that list, together with their recommendations for corrective action. At which point countries will simply go about their business as before. Does anyone really think that the debate between the Republicans and Democrats in the Congress over the US budget deficit will be fundamentally altered, or even affected on the margin, by advice proffered by the IMF and America’s G20 partners? Or that the debate in China between export interests pushing for the maintenance of the current exchange rate and the advocates of greater flexibility would be fundamentally reshaped?
For those of more optimistic bent — and for officials who do not wish to believe that they are wasting their time — the argument must be that attitudes and outlooks will evolve as a result of this process. Under the Multilateral Consolidation Initiative, consultations were ad hoc, countries could agree to disagree, and that was it. The G20 process, in contrast, is ongoing. Over time, as analyses are repeated and shared, officials will gravitate toward a common diagnosis of the problem and its solution. American officials will be socialised into understanding that their monetary and fiscal policies are a problem for the rest of the world. Chinese officials will come to appreciate the problems that their country’s chronic current-account surpluses pose for other countries. They will come to share a diagnosis of what needs to be done. Importantly, they will also be able to sell it to their political colleagues and constituents at home.
There are two things to say about this. First, it is a decidedly optimistic interpretation about how outlooks and attitudes evolve. It presumes that dialogue, mediated by a fair broker like the IMF, leads with time to a meeting of the minds rather than a rupture. It presumes that exercises in marriage counselling are an opportunity for introspection and bring out the rational instincts of the participants, rather than simply providing another venue for bickering and recrimination.
Second, it is a caution against unrealistic expectations. Little can be expected to result from the G20 process in terms of substantive policy changes in the short run. But it is the process itself that is important. It is the process that will deliver more extensive international cooperation in the long run.
The key question is how long is long. It would be unrealistic to suppose that a strong convergence of national perspectives — not just at the level of leaders but also other national politicians and their constituents, on matters as contentious as debts and deficits, external as well as domestic — could occur in less than, say, five years. This is a reminder that the G20 process is unlikely to protect us from the risks posed by the current constellation of global imbalances and their disorderly unwinding. This means that members need to take other steps to prepare themselves for this eventuality.
Editor’s note: This was written for the German Marshall Fund of the US; the original can be downloaded here.