World 3.0

by Pankaj Ghemawat

  Thus, the debate on the overall impact of capital account liberalization rages on. But from the World 3.0 perspective, what's more interesting is the evidence on the benefits and risks involved with specific types of international capital flows. Foreign direct investment (FDI)—foreign companies buying, setting up, or reinvesting in businesses in a country—tends to represent a long-term commitment since such decisions are typically driven by long-term strategic considerations.30 As a result, while the amount of new FDI can fluctuate quite a bit—especially around waves of mergers and acquisitions—investments are seldom withdrawn quickly and, from the host country's perspective, have the advantage of not requiring regular interest payments. If host markets do falter, such firms tend to be relatively well positioned to reorient production to exports rather than pulling out. And more broadly, FDI helps transfer knowledge and information as well as capital, and functions, like trade, as a channel for product market integration with the prospect of ADDING value in ways that go beyond the ones listed here.

  Opening up to international stock market investment (portfolio equity flows) has also been shown to have positive effects. Equity market liberalization has been linked to faster growth.31 And increasing financial integration, as measured based on the convergence of equity risk premia, has also been associated with less volatile growth.32

  International portfolio debt and bank loans, however, are substantially more problematic. When host country conditions deteriorate, such “procyclical and highly volatile” flows don't simply shrink: they are susceptible to reversing course and making a mad dash for the exits precisely when a country needs them most.33 Thus, “there is a systematic empirical link between exposure to short-term debt and the likelihood (and severity) of financial crises.”34 And heavy reliance on portfolio debt has been identified as a likely contributor to developing countries' inability to smooth consumption via international risk sharing.

  To summarize, opening up to international capital flows is a mixed bag. Thus, the conventional wisdom today is that “countries should liberalize trade in goods before trade in financial assets.”35 When countries do open up their capital markets, they can still favor some flows (FDI, equity) over others (debt), and retain the policy flexibility to manage extreme bouts of volatility. As Paul Krugman put it, “just as the right to free speech does not necessarily include the right to shout ‘Fire’ in a crowded theater, the principle of free markets does not necessarily mean that investors must be allowed to trample each other in a stampede.”36 If capital is like air to a market economy, its supply can't be left entirely in the hands of financial markets that, because of informational and other imperfections, have a history of “manias, panics, and crashes.”37 The wake of a global financial crisis should be an easy time to secure agreement—except from hardcore “free-marketeers”—on that basic proposition.

  Food Fights

  Losing money is scary, but the prospect of not having enough to eat is far scarier. Roughly 1 billion people were undernourished in 2009, one-sixth of humanity, a number that, disturbingly, has risen since the mid-1990s after declining steadily for decades.38 Globalization often takes heat for this, in large part by allegedly contributing to food price volatility. What the evidence actually indicates, though, is that impediments to food trade cause much of the trade-induced volatility, not exposure to free international markets. Only 18 percent of wheat, 7 percent of rice, and 10 percent of coarse grains (including corn) were traded across borders in 2008,39 and these three staples account for 60 percent of the world's caloric intake.40 Overall, agricultural trade constituted only 8.5 percent of total merchandise trade (down from 12.2 percent in 1990).41

  One reason is protectionism. Developed countries in particular maintain tariffs, subsidies, quotas, production “set-asides,” and so on. Many readers have heard the alarming statistic that farmers in the EU received a $913 subsidy in 2000 for each dairy cow they raised, while Japanese farmers got $2,700.42 Billions of people live on fewer dollars per day.

  To see how limited trade contributes to food price volatility, consider rice markets. The proportion of rice traded across borders rose from 4 percent in the 1980s to 7 percent in the early 2000s—a deepening of trade accompanied by falling price levels and declining price volatility.43 Rice trade remained thin and unstable, though. From early 2007 to mid-2008, international rice prices tripled. Why? Following a run-up in wheat prices and a poor wheat harvest in India, India and Vietnam (which together with Thailand account for 60 percent of rice exports) placed restrictions on rice exports (see figure 7-3). This triggered a cycle of panic buying, hoarding (speculation), and more export restrictions until prices overshot what fundamentals could possibly justify. Then, after some reassurance from Japan, prices fell almost as fast as they had risen.44 Sound familiar? A similar cycle appeared to be under way in 2010.

  Figure 7-3: Rice price trend and major trade-related events during the financial crisis

  Source: Derek D. Headley, “Rethinking the Global Food Crisis: The Role of Trade Shocks,” International Food Policy Research Institute discussion paper 0958, March 2010.

  The impact of export restrictions in the rice market spike suggests that open markets would have led to more stable prices, improving food security. Other research supports this view. An analysis of wheat prices in Zambia indicates that in a bad year, with harvests 30 percent below normal, prices might increase 30–40 percent if imports (from within the region) were allowed, but more than 150 percent if they weren't!45

  Despite such evidence, some continue to worry that foreign pressure on food prices will be exacerbated by population growth and global warming. However, there are actually good historical and scientific reasons to be optimistic that our planet can sustainably feed the 9 billion population projected by the mid-twenty-first century. According to Matt Ridley, from 1968 to 2005 alone, the amount of cereal grains produced from the same amount of land doubled, and “since 1900 the world has increased its population by 400 percent; its cropland area by 30 percent; its average yields by 400 percent and its total crop harvest by 600 percent.” In Ridley's view, the natural limit is about 100 square meters of cultivated land to feed one person. Based on the yields of the most important food crops, he estimates that we now use about 1,250 square meters to feed one person, down from 4,000 in the 1950s, and still far enough above 100 to imply there's still a huge scope for improvement.46 Furthermore, he believes—I am not so sure of this—that global warming would probably improve agricultural yields and food supplies.47

  So trade can reduce food price levels and volatility. While pressure to keep a lid on domestic prices sometimes creates incentives to restrict trade, basics such as enhancing market transparency, hedging, maintaining buffer stocks, and so on are usually better policies to promote food security. And macro trends like population growth and climate change certainly don't imply we'd be better off with isolated national food markets.

  Risking Integration

  Now that we've looked broadly at how macroeconomic volatility has trended as integration has increased, and drilled down on capital and food markets, let's come back to the fundamental relationship between integration and risk. In the complex real world, we can't slam down the gavel and say definitively that integration always makes life more dangerous, or that it always makes us safer. I deliberately juxtaposed capital and food flows to make the point that reducing risk sometimes means curtailing flows (e.g., short-term debt) but in other cases calls for freeing them up (e.g., food). Contrasting these two examples more explicitly calls attention to factors that are generally useful to consider in analyzing the broad relationship between integration and risk.

  Begin by making sure that your policy objectives are clear. Food is critical to sustaining life, and shortages and price spikes cause hunger. Sudden capital flow reversals are also life-threatening to an economy, but foreign capital inflows are seldom essential, particularly when one considers that developing countries typically invest just over 20 p
ercent of national income and are rarely able to borrow more than around 5 percent of national income.48 Thus, availability and price stability are most critical for food, while avoiding sudden outflows is most important for capital.

  Then, think about the nature of volatility in these markets and how it relates to informational imperfections. Food production fluctuates due to natural phenomena such as weather. Weather is a “risk” in the terminology introduced by Chicago economist Frank Knight in his famous 1921 book, Risk, Uncertainty and Profit, meaning that its outcome is unknown but it does conform to known probability distributions. Food export restrictions embody more of what Knight calls “uncertainty” in the sense that outcomes aren't based on a known probability distribution. Informational problems associated with uncertain trade barriers are likely to cause food prices to veer away from fundamentals to a greater extent than those rooted in more easily quantified production risks. The volatility of short-term capital flows is likely to be even more problematic since it is rooted in sentiments rather than in a specific natural production risk at all, and will tend toward the hot and cold pattern of overshooting that is common to financial markets.

  Furthermore, compare the likely benefits of diversification in these two markets. Diversifying food sources across locations with different weather patterns yields obvious benefits. In capital markets, there are also diversification benefits associated with bridging economic distance, but “herding” reduces the benefit of diversification for borrowers, as we saw in the cases of sudden stops.

  Based on this analysis, it should be fairly clear why ensuring open markets for food grains reduces risk while freeing up short-term debt flows can raise it. Furthermore, even when a flow does appear to offer big benefits, it's important to think about risks rooted in the relationships between the specific countries involved. A short digression beyond capital and food to energy—another essential commodity that requires management of cross-border risks—illustrates the point. Denmark exports wind energy to Norway when the wind is blowing, enabling Norway to cut the flow across its hydroelectric plants and store energy for later. When winds are light, the flow reverses, and Denmark imports hydroelectric power from Norway. In this case, relations between the countries involved are such that little risk of unexpected electricity cutoffs exists. On the other hand, the proposed natural gas pipeline from Iran to India across Pakistan seems to raise lots of issues in this respect. Similarly, if there are public concerns about a flow, that should be a signal to proceed cautiously, if at all: when the going gets tough, institutions come under pressure and rules may be broken.

  Toward a Safer World

  The last section provided examples of the potential for risk reduction through international diversification that is effectively ignored in World 1.0 and World 2.0. World 1.0 is integration-averse and seeks safety behind closed borders. World 2.0, at the other extreme, also fails to emphasize the potential benefits of risk-pooling because it ignores cross-country differences; in addition, it suggests no protections against the real risks described in this chapter. By tuning us in to differences and market failures, World 3.0 flags both the potential to reduce risk through integration and the challenges of combating the risks that integration and interdependency can themselves generate. It is time to discuss how to deal with those challenges.

  As a starting point, pursue diversification. Any investor knows it's too dangerous to hold only one stock. But World 1.0 tells us it's better to depend on just one country—home—than to diversify. That's awfully risky. I don't suggest going to the World 2.0 extreme and trying to spread a country's exposure evenly across all other countries. Watch out, for example, for exposure to countries with which your home country has hostile relations. But all else equal, insurance through risk pooling does have some value.

  That said, diversification is far from sufficient to manage risks in a semiglobalized world with informational imperfections. As highlighted in the previous section, certain kinds of cross-border flows can amplify systemic risk—which diversification doesn't alleviate—or create new dangers. And since waiting to deal with such problems until they become pressing is typically less effective and more expensive, we need to think of prudential measures to take to counter cross-border risks, particularly capital risks, rather than simply reacting on an ad hoc basis as problems come up. I summarize some of the possibilities under the rubric of the ABCs: alarms, breakers, and cushions. Alarms are early warning systems, breakers curb contagion, and cushions soften the blows, particularly for the most vulnerable.

  Setting effective alarms often begins with bolstering information flows: to manage risks, you have to know your exposure. The fact that initial estimates of Swiss banks' exposure to Greece were off by 95 percent—a matter of some $60 billion—due to a classification error is staggering.49 What makes it worse is that we are talking about the Swiss banking authorities! Discovery of the error reduced (non-Greek) European banks' estimated exposure to Greece by nearly one-quarter.

  To use alarms effectively, define specific exposure metrics and threat levels. Alarms may simply be advisory but can also, especially when the degree of threat is acute, trigger specific responses, as in the case of U.S. Homeland Security's terror alert level. For capital flows, an obvious focus given the discussion in this chapter and the next is on the absolute value of countries' current account balances. Based on the historical data summarized in figure 7-4, start paying attention when they reach 3 percent of GDP and start getting worried when they exceed 4 percent. Other possible metrics include foreign debt as a proportion of GDP, and the measures of cross-border capital mobility calibrated in chapter 2. And threat levels often need to be tuned to country characteristics.

  One specific action that serious threats can trigger is the activation of various kinds of breakers to curb market excesses. Joseph Stiglitz provides the following analogy in a paper subtitled “Why Full Financial Integration May Be Undesirable,” which also reminds us that some of the same ideas about managing systemic risk should also apply to energy, transportation, and information infrastructure, among other areas:

  With an integrated electric grid the total capacity required to limit the probability of a blackout to a particular level can be reduced. But a failure in one part of the system can lead to system-wide failure; in the absence of integration, the failure would have been geographically constrained. Well-designed networks have circuit breakers, to prevent the “contagion” of the failure of one part of the system to others.50

  Figure 7-4: Size of net capital flows since 1870, for selected countries (average of absolute values of current account balances as % of GDP)

  Countries covered: Argentina, Australia, Canada, Denmark, France, Germany, Italy, Japan, Norway, Sweden, United Kingdom, and United States.

  Sources: Maurice Obstfeld and Alan Taylor, “The Great Depression as a Watershed: International Capital Mobility Over the Long Run,” NBER Working Paper 5960, March 1997 (data before 1997), World Bank, World Development Indicators (data after 1997).

  In a general capital market context, limits on daily movements in prices of individual stocks or broad market aggregates and suspension of trading when those limits are reached provide an example of circuit breakers. In a specifically cross-border context, capital controls may have to be considered as a moderating influence, especially by small economies, although they do have undesirable side effects. More broadly, figure out which of the dominos (as in the cartoon at the beginning of this chapter) should be kept farther apart or be reinforced or be removable in times of crisis to keep negative chain reactions from running amok and causing systemwide as opposed to isolated damage.

  When safety mechanisms fail and trouble does develop, cushions can help soften the blow. Maintain strategic reserves. Put appropriate hedging and insurance arrangements in place. Prepare policy responses ahead of time to a range of macroeconomic contingencies. Remember the special role of the government: when disaster strikes, it's still national governments that
get the call, and so they need to be in a position to meet basic requirements such as food security. And perhaps most importantly, build redundancy into critical systems. Thus Nassim Taleb, the author of The Black Swan, chides economists for failing to respect nature's preference for redundancy and the robustness that it implies: “An economist would find it inefficient to maintain two lungs and two kidneys.”51

  More controversially, these ABC's might be augmented with a D in the form of dampeners. Breakers are designed to be applied only in the event of trouble, but there are some indications that with widespread speculation and high frequency trading, it might be necessary to “throw some sand in the wheels” of market operations on an ongoing basis to make them function better. This was the idea behind Nobel laureate James Tobin's proposal for a small tax on all international currency transactions, intended to curb waves of speculative capital flowing in and out of countries while leaving healthier longer-term flows basically unaffected.