dani rodrik is the Ford Foundation Professor of International Political Economy at Harvard’s Kennedy School of Government. His book, Straight Talk on Trade: Ideas for a Sane World Economy, will be published this year by Princeton University Press.
Published October 20, 2017
Electorates around the world were told not only that globalization was inevitable, but also that it necessarily took the particular form they were witnessing. The nation-state, it was said, was the enemy of globalization, and therefore had to get out of the way. Globalization required ever-stronger global rules mandated by trade agreements, multilateral organizations and international networks of regulators. But not to worry: it would promote economic progress and political harmony, even if not for everyone right away.
None of this was quite true. There is nothing inevitable about advancing economic integration, nor about the route that globalization takes if it does move forward. And contrary to conventional wisdom, nation-states are absolutely essential to globalization because they provide the public services ranging from law enforcement to macroeconomic stabilization that are needed for open markets to thrive. By the same token, global governance is largely superfluous: proper trade, financial, monetary and regulatory policies required to sustain an open world economy do not require much coordination when governments do their jobs well.
Consider, too, the idea that nations needed to shape up to benefit from integration. This transformed globalization into an end rather than a means. The right and the left differed on the regimen needed. The right emphasized getting the investment environment right: cutting red tape, for instance, and reducing corporate taxes. The left talked about investment in skills, education and infrastructure. But in both cases, globalization was assumed to be the goal, not the route to the goal.
Now, globalization certainly has generated benefits. The professional, managerial and capitalist class of the advanced economies — those who were already doing quite well — gained tremendously. Just as important, so did hundreds of millions of underemployed poor people in China and elsewhere (but mostly in Asia) who found jobs in production for export markets. But look closely at China and other East Asian countries that did so well and you discover that they played the globalization game by different rules. They opened their markets only partially, governing the pace and impact of economic integration with interventions ranging from subsidies for favored industries to controls over cross-border capital flows.
The populist backlash epitomized by Donald Trump’s election victory but brewing since the 1980s has convinced even globalization’s cheerleaders that some things must change. Elites now concede that globalization produces losers as well as winners. But the correct response, they argue, is not to change course with globalization; it is to ensure that the losers are compensated.
The logic of sustaining an open economy by compensating those who end up with smaller slices of the pie is impeccable. That’s how European nations, with their extensive safety nets and generous social benefits, integrated into the world economy. To this day, despite rising populism, international trade is not a very contentious issue in Europe. Anti-globalization ire focuses not on Chinese or Mexican exporters, but on faceless bureaucrats in Brussels and Frankfurt — and, of course, on immigrants. The United States, too, could have moved aggressively to compensate dislocated workers in the 1990s, when it opened its economy to imports from Mexico, China and other low-income countries in a major way. Instead, under the sway of market fundamentalists, the United States let the chips (and workers) fall where they may.
By now, the compensation approach has been tarred as “burial insurance.” The trade adjustment assistance programs that are habitually tacked on to trade agreements have provided inadequate aid — and to just a sliver of the affected population. That is partly by design: politicians have little incentive to implement strong compensation programs once trade agreements are approved.
American workers have been the weak party in the bargain all along — if they’d had enough clout to obtain a robust safety net, they would have had the clout to reshape trade agreements in worker-friendly ways in the first place. Where compensation has worked — as in Europe — it has been part of a broader settlement between business and organized labor in which workers accept the sort of labor-market instability that is part and parcel of open trade in return for the welfare state.
In my view, if globalization is to be saved, it won’t be by renewed promises of handouts or training in computer programming and other skills. What’s needed is a significant change in direction. And fortunately, it really would be possible to preserve globalization’s economic benefits and make it fairer at the same time. There are many promising paths that prevailing narratives have led us to overlook. In particular, we need to rebalance policies toward global economic integration in three areas: from capital and business to labor and the broader society, from global governance to national governance, and from areas in which overall economic gains are small to where they are large.
From Capital to Labor
The benefits of globalization are distributed unevenly because our current model of globalization is built on a corrosive asymmetry. Trade agreements and global regulations are designed largely with the needs of capital in mind. The interests of labor — good wages, decent work environment, employment security, voice in the workplace, bargaining rights — are scarcely paid lip service. The implicit economic model is based on the trickle-down effect: the benefits to investors will spill over to society in the form of better jobs and tax revenues.
Hence globalization has taken a skewed form. It’s all about reducing the costs of doing business across borders and facilitating cross-border capital flows. The World Trade Organization and the International Monetary Fund command all the attention and carry the big sticks, while the International Labor Organization is merely a talking shop.
Trade agreements are driven overwhelmingly by a business-led agenda. Protection of property rights of investors, even when rights are nebulous (as in the case of intellectual property rights), is the first priority. Investor-state dispute settlement, which allows businesses to sue national governments before international arbitration tribunals, has become ubiquitous. It is purportedly about ensuring that governments live up to their own laws and commitments and do not take advantage of foreign firms. But why is the privilege of seeking redress through international arbitration granted only to firms and foreign investors? Why not have similar international tribunals in which labor (or consumer) groups can challenge violations of core labor rights or consumer safety laws by trade partners?
A striking reflection of the asymmetry is that capital and business are free to move across national borders, while labor is not. We regard this as a natural feature of the world economy, even though it is a consequence of our own policy choices and there is nothing natural about it. During an earlier era of globalization, at the tail end of the 19th century, people were relatively free to move to countries where there were better economic opportunities — and they did so in great numbers.
Moreover, free movements of capital have not always been in fashion. At the end of World War II, economists and policymakers were unanimous in regarding capital mobility — anything other than long-term direct business investment — as inherently destabilizing and inimical to macroeconomic management. By the 1990s, however, international capital mobility had become the global norm, enshrined in the practices of the OECD and the IMF, while labor mobility remained highly restricted and regulated by vastly different domestic laws.
Differential mobility produces a wide range of distributional effects that advantage capital and disadvantage labor. Most obviously, investors and highly skilled professionals can benefit from higher returns on the other sides of borders; the vast majority of workers cannot. And this in turn has implications for bargaining between labor and capital.
The threat of leaving or outsourcing can force labor to take a pay cut or a smaller wage increase. Hence the global trend toward lower labor shares of total income. Capital mobility makes it more difficult to tax capital — and on the flip side, creates greater incentives to subsidize it in order to bring it home. Hence the global reduction in corporate tax rates and competition for investment. Finally, differential mobility means that economic risk is borne disproportionately by the immobile factor, labor. Economic downturns mean long spells of unemployment and wage cuts, while investors are partially protected by their global diversification.
To right the capital-labor imbalance, labor must be given an equal say in setting the rules of globalization. In practical terms, this requires reconsidering which multilateral institutions set the agenda of the global conversation and who sits at the bargaining table when trade agreements are negotiated.
Giving labor a bigger say would not necessarily produce greater worker mobility across borders; workers in advanced and poor nations have conflicting interests. But it might at least lead to the rejection of the current norm in which capital is globally mobile while labor is not. I believe this norm should be replaced by the idea that capital and labor mobility must go hand-in-hand, with the degree of overall mobility determined by other policy considerations. At present, capital is too mobile, while labor is not nearly as mobile as it should be. So such a rebalancing would be a good thing — and for efficiency as well as for distributional reasons.
From Global Governance to National Governance
While national economies are linked in a variety of ways, the world economy is not a global commons and therefore does not need global governance in order to be managed effectively. Indeed, most failures in the world economy are rooted in failures of domestic governance. Therefore, the best fix for the global economy is better domestic governance.
This assertion certainly runs against the tide, but does follow directly from the economic doctrine that underpins globalization: the mutual gains from trade. As David Ricardo, the influential British political economist, showed long ago and generations of economists have professed since, nations trade because it is in their own interest. Comparative advantage — the market-driven process by which economies specialize in what they do or make relatively well — generates gains from trade for both parties. The point of opening up to trade is not to confer economic benefits on other nations, but to improve your own lot. Of course, trade generates losers in a variety of ways as production is dislocated. But trade is not different in that respect from technological progress. Well-functioning societies find ways of reaping the overall benefits by bringing the losers along.
Consequently, failure to adopt open-trade policies must be due either to complications that undermine the standard Econ 101 argument for the gains from trade, or to domestic policy failures. The former arises when economic or social considerations render open trade less than optimal. For example, society may value environmental amenities (say, clean air) or distributional outcomes (maybe generous blue-collar wages) that would be undermined by free trade, and it may not be possible to offset the losses by other means.
Consider, too, that a developing country may benefit from sheltering infant industries from foreign competition before it has developed competitive means of production. In such instances, the rejection of free trade does not undermine productivity growth from either domestic or international perspectives. Global agreements have no business telling these countries to open up.
In the second case, domestic policy failure, there is a genuine problem. But the problem lies in domestic politics — the excessive political influence, perhaps, of entrenched interests like dairy farmers or automakers — and not in the absence of proper global rules.
Empowering global governance in such circumstances may or may not serve to manage such rent-seeking. Global rules sometimes act as a counterweight to protectionist interests. But it is equally likely that the global rules will be written and administered by the very same special interests that dominate domestic policymaking as well. Think of the role of big banks in setting global minimum capital standards or pharmaceutical companies in writing global patent rules.
Similar reasoning applies in other realms of international economic policy. Appropriate prudential regulation for financial markets and sound fiscal and monetary management are good for the home economy. When policy fails in these areas, it is usually not because one country is putting its interests above others (i.e., beggar thy neighbor), but for beggar-thyself reasons. Global governance sometimes can make things better by giving good-government types an advantage (as in, “If we don’t do the right thing, the World Bank won’t lend us money”). But there is no avoiding the fact that the problem is with domestic policy and politics.
There are exceptional circumstances, of course, where national and global interests do not coincide.
Countries do sometime pursue mercantilist policies that are of the beggar-thy-neighbor kind. Thus contemporary Germany’s suppression of wage growth in the name of boosting exports makes it that much harder for less-productive Eurozone economies to recover from the recession. And China’s (mostly indirect) subsidies to exporters in order to accelerate growth did sometimes come at the expense of efficient producers in trade partners. But few of our major contemporary problems in global trade and finance are of this type. Misguided protectionism, excessive fiscal austerity, inadequate financial regulation and poor property-rights protection are all beggar-thyself policies.
Economists often assume that global rules are more in control of technocrats and less prone to political manipulation than national ones. But even if that is true, it is not clear this is an advantage. In fact, many of our existing modes of global governance — multilateral agreements on everything from investment regulations to codes of corporate responsibility — raise troubling questions. To whom are these mechanisms supposed to be accountable? Where do these clubs of regulators, international non-governmental organizations or large firms get their mandates? What ensures that the voice and interests of those who are less globally networked are also heard?
In democratic nations, the electorate is the ultimate source of policy mandates and elections are the ultimate vehicle for accountability. If you do not respond to your constituencies’ expectations and aspirations, you are voted out. While plainly flawed, the democratic state is tried and tested. Its global counterparts are too distant from direct accountability and hence lack legitimacy.
None of this implies that there is no role at all for global governance. But in rebalancing globalization toward national control, we must understand that the highest priority of global arrangements ought to be making the nation-state work better, not weakening it. Correspondingly, the appropriate role for global institutions is to enhance key democratic norms of representation, participation, deliberation, rule of law and transparency — without prejudging policy outcomes from the perspective of the need to harmonize cross-border policies.
Global conversations about the spillover effects of domestic macroeconomic or regulatory policies are sometimes useful. Such conversations can be helpful in reducing international misunderstanding about the objective of policies and sometime in establishing new behavioral norms. They can also facilitate deal-making in cases in which it’s worth enough for one country to pay off a second to stop acting in ways detrimental to the first — say, by providing technology to reduce water pollution that ends up downriver.
But the presence of negative spillovers alone does not amount to a case for global restraints on domestic policy space. Nowhere is it written that the economic interests of other nations ought to take precedence over socio-economic benefits to the home nation. Global arrangements should be focused on enhancing democracy at home rather than globalization per se.
From Places Where Net Economic Gains Are Small to Where They Are Large
The argument for openness is that it produces efficiency gains and hence an expansion of the total economic pie. It would stand to reason that negotiations on lowering trade barriers should focus where these gains are largest and domestic political costs and adverse distributional effects are smallest. Viewed from this perspective, our conventional agenda of globalization looks very curious. A lot of political capital is wasted on efforts that produce little net gain, while areas in which net gains could be huge remain untouched.
To see this, it helps to start with a basic principle of public finance. A government-imposed restriction on trade is like a tax — a tax on imports. And the efficiency cost of a tax rises disproportionately, with the square of the rate of the tax. A tax that is twice as large is four times as harmful to the economy in terms of efficiency; a tax that is four times as large is 16 times as harmful. What this means is that a reduction in a barrier that is high produces much larger gains than an equal reduction in a barrier that is already low.
Redistribution of income and wealth, by contrast, creates linear effects. An equal reduction produces similar effects regardless of the initial height of the barrier that is impeding trade.
Now put these two realities together. The larger the amount of redistribution that is generated by trade opening per dollar of efficiency gain, the smaller the barriers that policymakers are going after. I have called this ratio of redistribution per net gain the political cost-benefit ratio of trade liberalization. After almost seven decades of trade negotiations and agreements, most barriers to trade in industrial goods and agricultural products have come down substantially. Chipping away further at these barriers thus produces modest net gains in efficiency accompanied by disproportionately large changes in income and wealth distribution.
This might seem like a theoretical point with little practical consequence. Not so. Consider the effects of the North American Free Trade Agreement on the U.S. economy. When the agreement went into effect in 1994, U.S. barriers to Mexican imports were already quite low. Not surprisingly, in light of the theoretical considerations above, subsequent empirical analyses have found very small efficiency gains for the United States: a sophisticated academic study by the economists Lorenzo Caliendo of Yale and Fernando Parro of Johns Hopkins uses all the bells and whistles of modern trade theory to generate an estimate of only a 0.04 percent gain in economic efficiency. You didn’t misread that: four-hundredths of one percent!
Nonetheless, Nafta shuffled a lot of income within the United States. The most careful analysis to date has been carried out by the economists Shushanik Hakobyan of Fordham and John McLaren of the University of Virginia. They found that an “important minority” of U.S. workers suffered substantial income losses. A high-school dropout in heavily Nafta-impacted locales had eight percentage points slower wage growth over 1990-2000 compared to a similar worker in a region/sector unaffected by Nafta trade. Wage growth in the most protected industries that lost their protection fell 17 percentage points relative to industries that were unprotected initially.
These are very large effects — especially when compared with the meager net benefits noted above. Of course, there were plenty of big winners as well as big losers, which explains why many corporations were in favor of the agreement. But in light of numbers such as these, it is not difficult to understand why Nafta remains so politically charged after nearly a quarter-century.
As tariffs came down, trade negotiations have evolved from a focus on direct barriers at the border to indirect barriers linked to domestic regulations (in areas like services, standards, patents and copyrights). Today’s trade agreements aim at harmonizing these regulations or reducing their impact so as to make it easier for firms to operate across markets.
The costs to enterprises associated with regulatory divergence can sometimes be high. However, unlike the case of barriers at the border, there is no general theoretical presumption that reducing or harmonizing these regulatory barriers enhances efficiency. In many cases, the regulations exist to promote social welfare (such as enhancing consumer safety, internalizing environmental externalities or providing access for disadvantaged groups). There is no reason to believe that harmonizing them for the purpose of expanding trade enhances overall welfare. The truth behind such negotiations is that they serve the purpose of specific interest groups that have managed to capture the process.
The cost-benefit ratio of financial globalization looks even worse. Capital mobility has not only increased income inequality around the world but it has also increased the incidence and severity of financial crises. Its vaunted benefit — promoting growth by transferring savings from rich to poor nations — has not materialized. Indeed, capital often flows uphill from poor country to rich. Yet global officialdom spends inordinate effort on harmonizing capital adequacy and other prudential regulations on the discredited presumption that financial openness produces large efficiency gains.
So where are the potential gains from globalization today? The political cost-benefit ratio logic I outlined earlier suggests an easy answer: where the barriers are really high. And they are nowhere higher than in cross-border worker mobility. Michael Clemens, an economist at the Center for Global Development, Lant Pritchett, the Harvard economist, and Claudio Montenegro of the World Bank calculated the height of the implicit barriers by estimating the income gains that would hypothetically accrue to a worker who moved from a developing nation to the United States. They found, for example, that a Pakistani worker would increase his income more than six-fold. Putting it in terms used to measure barriers to goods and services, this means that in an open market, it would take a tax or tariff of 500 percent on worker income to deter voluntary labor movement — vastly higher than anything in trade.
I have long argued for a temporary worker-visa plan that would take advantage of such unexploited gains. It would be administered bilaterally on the basis of specific home-country quotas. To maximize home country benefits and spread the gains around, the visas would be for a fixed period — say three to five years. The visas would not entail a path to citizenship, although guest workers would have the full protection of the host country’s labor standards and regulations.
A mix of sticks and carrots might be employed to ensure the bulk of workers do choose to return to their home countries when their visas run out. For example, a portion of guest workers’ pay could be held in forced saving accounts, to be returned only upon repatriation. The quotas of worker-exporting countries could be adjusted in relation to their success in attracting their workers back home. This would give home countries an incentive to provide repatriation inducements, just as they now woo foreign investment and skilled expatriates.
Expanding worker mobility across borders in a negotiated, managed manner would produce a large increase in the size of the economic pie, both globally and nationally. Of course, it would also have some redistributive effects, especially in the short run. It would likely hurt some unskilled native workers in the rich nations. But the political cost-benefit ratio logic suggests that the redistribution we’d get (per dollar of efficiency gain) would actually be comparatively small given the implicit height of current labor-mobility barriers. In addition, the guest workers would be employed under domestic labor standards, rather than home-country standards that are likely far weaker. This would remove an important source of concern in host economies with regard to unfair trade and “social dumping.”
A second area where the efficiency gains from increased openness would be comparatively large consists of reforms that would enhance the legitimacy of the world trading regime. For too long, trade negotiators have operated in an “exchange of market access” mindset: you open your market, and in return I will open mine.
But openness is good for each (in terms of efficiency), regardless of what the other does. Even if one ignores the mercantilist orientation that seems curious two centuries after David Ricardo campaigned to eliminate tariffs on wheat, this approach has run its course. The binding constraint on global integration today is not that it is insufficiently open. It is that too many people believe it is managed by plutocratic elites for their own benefit rather than for the benefit of the majority.
This suggests the need for a hard look at trade agreements and the World Trade Organization regime to remove intrusive rules that limit the ability of nations to respond to domestic needs. For developing countries, it means lifting restrictions on industrial policies, patents and copyrights and capital-account management. For advanced economies, it means allowing greater scope for remedies against social dumping and unfair trade. Overall, it requires a transition in global talks from “exchange of market access” to “exchange of policy sovereignty.”
It really would be possible to design a global economic system that is simultaneously attentive to the needs of developing and advanced economies. American progressives are unnecessarily stymied by this issue, worrying that trade policies prioritizing the interests of American workers would necessarily harm much poorer workers in the developing world. In fact, rich and poor countries alike could benefit from a world trade regime with a lighter touch. Some of the most impressive growth miracles of the post-war era — Japan, South Korea, Taiwan — came to pass at a time when the WTO did not exist and multilateral constraints on trade barriers were limited.
There is, of course, no guarantee that governments would use the maneuvering space they are given to good effect. But at least they would not have the excuse that “globalization made me do it.” Domestic policy failures would be exposed for what they are.
Eyes on the Prize
We should not reject globalization, but save it in a form that works better for more people. Economic integration has overshot in some areas, such as financial globalization and regulatory harmonization. It has not gone far enough in others, such as international labor mobility. The debate we should avoid is whether globalization per se is good or bad. The real question is how to rebalance it to give excluded groups greater voice, reconstruct social compacts at home and focus our global negotiations on areas where the potential economic gains are still really big.