The interconnected world is not new. We have been living in it for some time. For half a century, exports, international capital flows, and foreign direct investment have been growing much faster than GDP, binding economies ever more tightly together. In many ways, this process has been a tremendous force for good. My own country, India, is testament to that. Integration with the world contributed to growth, lifting hundreds of millions of people out of poverty. But in the past few years, we have seen that these linkages can also have tremendous costs: developments in other countries, over which we have no control, can without warning plunge countries into difficulty or even crisis.

2. So, we need to find ways to maximize the benefits of globalization while minimizing its costs. To do this, there are two basic imperatives. First, we need to better understand the ways in which we are linked together. And then, we need to translate this understanding into coordinated policy action. That is, we need to consider ways in which countries can impose policy discipline on one another’s behaviour, to minimize the risks that their policies could pose for others. I would like to focus my brief remarks today on these two tasks: understanding the nature of linkages and doing something about them.

Understanding Linkages

3. We cannot begin to impose policy discipline on countries until we improve our understanding of the linkages amongst economies. Recent experience shows that our understanding of the nature of these linkages is seriously deficient. In fact, standard econometric models still suggest that shocks in one country have very small effects on other countries. They do that because they assume that trade is the primary channel for transmitting these shocks across countries, and trade moves in line with GDP.  Both these assumption are flawed. As we all know, the 2008 crisis originated in the housing sector which is a quintessentially non-traded good.

4. In addition, we saw during the crisis that the financial channel is tremendously important, in ways that are still poorly understood. For example, after Lehman collapsed, interbank rates in India soared, by much more than in many other emerging markets, even though India’s banking system is relatively closed.

5. Finally, there is another channel, little analysed and even less understood, arising from the global nature of modern industries. For example, in 2001 the collapse of the IT industry was a world-wide phenomenon, one that was felt even in India, with a magnitude that would never have been predicted by trade or financial equations. Right now, we are worrying about the implications of Japan’s difficulties for global supply chains. But we have no framework for analyzing these effects. So, further research into linkages is needed, and the IMF is well-placed to lead such an effort.

Dealing with Spillovers

6. Once we gain a better understanding of the way in which economies are linked together, we can begin to devise better ways of dealing with them. But this will also be a difficult task.

7. Consider first the relatively straightforward case where bad policies in one country can have negative consequences for the rest of the world. The most obvious example is poor financial regulation, which was one of the root causes of the 2008 global crisis. In principle, it should be easy to deal with this problem since bad policies are harmful for all. Yet as we are seeing, there are immense difficulties in agreeing what good financial regulation entails. Perhaps this is not surprising. After all, it was not too long ago that the systems in advanced countries were considered “best practices”, which Emerging Market Economies (EMEs) were advised to emulate. So, we now need some fundamental research into what good regulation really entails, and this research would do well to consider our total experience - not just that of advanced economies but also of EMEs.

8. Consider now a more complex case, where there are both positive and negative spillovers to analyse. A prominent example is the second round of US quantitative easing, QE2. From the perspective of EMEs, QE2 may have had significant negative externalities. After QE2 was announced in August 2010, the prospect of easy liquidity in the US seemed to prompt a large increase in capital flows to EMEs, threatening domestic price and financial stability. It also seemed to contribute to rising global commodity prices, intensifying inflationary pressures.

This combination has put some EMEs in a policy bind, as rising inflation necessitates tighter monetary policy, but higher interest rates will only intensify volatile capital inflows, potentially putting more pressure on exchange rates and domestic stability. In India’s case, the concerns on this account currently are less acute since capital inflows are needed to finance our current account deficit. Yet even for us, the composition of the inflows remains an issue. About three-quarters of the current account deficit since 2009 has been financed by volatile capital inflows.

9. At the same time, it must be acknowledged that QE2 has also had positive externalities. It seems to have improved confidence in the US economy, especially by putting an end to deflationary fears. So, to the extent that it has contributed to a stronger and more durable recovery in the US, it has also benefitted EMEs, including India.

10. This case illustrates that before one imposes discipline on spillovers, it will be critical to assess their impact. First, one would need to identify the various external effects, quantify them, and find some way of netting them, to see whether the overall impact has been positive or negative. Then, one would need to balance these external considerations against the purely domestic benefits. Merely to state these issues is sufficient to illustrate the magnitude of the challenge before us.

11. Finally, consider the case where one country’s policy has a positive impact on the domestic economy, but a negative one on the rest of the world. A well-known example is when a country maintains an undervalued exchange rate. In India’s case, the rupee’s exchange rate has been virtually flexible in the past two years, as we have not intervened in the foreign exchange market. The small increase in reserves reflects various accruals, interest earnings and valuation changes. This policy has served the economy well, as it has allowed the exchange rate to serve as a buffer, depreciating to help the economy when it was weak and appreciating to reduce excess demand when it was strong. The policy has also minimized the danger that foreign inflows would be attracted by “one-way bets” on appreciation, or that domestic firms would borrow excessively from abroad without hedging their exposure. But at the same time, India’s policy means that it is subject to negative externalities from countries that maintain undervalued exchange rates, undermining our competitiveness in these markets and our efforts to contain the current account deficit.

12. The underlying question is about when and in what circumstances countries can intervene in the forex market to manage the exchange rate. In particular, is it appropriate to maintain an undervalued exchange rate for purely trade advantage? But equally, we recognize that there will be numerous difficulties. How can undervaluation be proved in a rigorous way? And where would one draw the line between an acceptable undervaluation, and a more serious one? The challenge is even more complex if the problem is defined in terms of current account surpluses, because there can be surpluses for good reasons (such as when the population is aging, leading the country to lend to younger, more dynamic countries that need to borrow) or bad ones, where the surpluses arise because of distortions.

Establishing a Code of Discipline

13. But let’s say that we establish a case for doing something about a spillover. The first step would be to develop a consensus on a code of discipline. As one IEO study noted :

It is now recognized that a world economy dominated by integrated markets, and with countries at very different levels of economic and institutional development, requires a system of global, yet non-binding, rules of conduct (“soft law”) that are internationally promulgated and nationally implemented….IMF surveillance has been seen as the instrument to disseminate the new sets of rules across the Fund’s near-universal membership, and to facilitate their implementation by member countries.²

14. Still, developing consensus on a code of discipline will be challenging. I have already alluded to the difficulties in reaching international agreement just in the specific field of financial regulation. It will be even more difficult to reach a consensus on broader economic policy. Indeed, the history of the past few decades is an object lesson in how problematic such a process could be. The first major code that emerging countries were expected to follow was the Washington Consensus of the 1990s. But not all countries benefitted sufficiently from these policies and at some point, there was a backlash.  The second wave of consensus came during the 2000s in the form of a plethora of international standards and codes. There was also a consensus that fiscal policy should aim at low levels of public debt, that monetary policy should target inflation, and that regulation should be separated from monetary policy, and used to preserve financial stability. The crisis has forced a rethinking on all of these. But a new consensus to replace the old one has not yet emerged, as the IMF conference held in March 2011 amply demonstrated.

Imposing Discipline

15. But assume we overcome this hurdle and reach a consensus on “the rules of the game”, at least in a particular area such as, say, exchange rate policy. How could the international community impose discipline on a country? Trying to do so through international agreements could confront two major challenges. First, countries could be reluctant to commit, because they are unsure how the commitments would affect the domestic economy. Second, the effectiveness of such commitments could be diluted if there is no true “buy in” by countries. The experience at the IMF with the 2007 Surveillance Decision, which proved impossible to implement is a cautionary tale. So, too, is the experience of the WTO, where countries have often evaded the spirit of trade commitments by resorting to protectionist measures on which no commitments were given, such as undervalued exchange rates, taxation and subsidies, labour laws, investment norms, intellectual property rights and environment laws.

Conclusion

16. So, how do we move forward? The way ahead is far from clear. But let me sketch out a road map.

17. The first step, as I mentioned, is to develop a framework for understanding the linkages amongst economies. This is currently being done through the IMF spillover reports, a most welcome initiative. But let me put the challenge squarely. We need to develop a proper framework for analyzing these linkages. And we cannot do this by using the existing trade equations, and attributing all the other spillovers to exogenous financial shocks. We need a deeper, truer, understanding of the channels and mechanisms that link our economies together. So, the IMF has an important opportunity to contribute to this understanding through its own research.

18. Second, we need to move toward a new consensus on economic policies, that is to say new norms against which country behaviour can be assessed. We have made considerable progress - though still not enough - on financial regulation. But we have only just started developing a new framework for macroeconomic policy, and defining the proper relationship between macro, growth, and financial policies. Again, the Fund could make a valuable contribution in this area, as well, both by advancing our theoretical understandings and by synthesizing the policy lessons learned by its member countries. As I mentioned, there is much for the world to learn by studying the experiences of the EMEs, as well as the most advanced countries.

19. Third, because our understanding of spillovers and best practices remains limited, it is far too early to think of reaching new formal international agreements on policy behaviour. Instead, countries will prefer to proceed through more exchange of information, peer review processes and ad hoc international agreements depending on the situation prevailing at any point of time. In other words, we will have to proceed by encouraging voluntary compliance by members, while also using peer group/multilateral spillover assessments as a soft pressure. The template for such an approach is the G-20 MAP process. The forthcoming reports on country imbalances, and how they might affect the other G-20 members, will give us an opportunity to exert peer pressure on countries whose policies may lead to dangerous spillovers.

20. Finally, we will need to come up with enforcement mechanisms on countries that create negative externalities for the global economy despite multilateral or peer assessment cautioning them about the spillover costs their policies generate for other economies. Since we are far away from a situation where sanctions could be employed, we will need to use incentives. For example, perhaps policy agreements could be reached whereby some countries abandon policies that are creating negative spillovers in return for other countries taking policy action with positive external spillovers for them. Designing such deals and getting countries to agree to them, however, will remain a major challenge.

21. In sum, before we can start imposing policy discipline, we will need to put in place the needed building blocks. We need to identify linkages, assess spillovers, and establish norms of behaviour. Then, once these norms are established, we can proceed to a code of discipline, with rewards and eventually perhaps even sanctions. It is a long road ahead. But the journey is not only worthwhile; in our interdependent world, it is necessary.

Annex

In the G-20 debate on spillovers, EME’s have emphasized that:

(b) the available policy space may vary across countries and over time (i.e. countries with managed capital account and exchange rate could enjoy greater policy space than countries with open capital account and flexible exchange rate),

(c) spillover management that may involve some sacrifice of domestic policy goals could be feasible only when every country pursues spillover-sensitive national policies, and the global governance structure would need to ensure that, using a mix of both persuasion and multilateral enforcement, and most importantly

(d) global safety nets should be strong enough with easier access norms to encourage countries to voluntarily adopt the multilateral norms.

¹Comments by Dr. Duvvuri Subbarao, Governor, Reserve Bank of India at the SNB-IMF Conference on The International Monetary System, Zurich, May 10, 2011.