A nice sentiment, surely. But is it good economics? My takeaway from China's experience is that it is, or it can be. There are lots of ways to spin China's policy of limiting the appreciation of its currency in order to promote export-led capital formation and growth. One story is simply that the policy amounted to a export subsidy: Purchasing power was withdrawn from Chinese workers and transferred to dollar and euro spending foreign consumers.

It's unmistakable that the policy "worked", in some sense. China's growth, along with the scale and pace of change in that country, have been remarkable.

I've mulled over the question of subsidy before. Simple economic reasoning suggests that subsidies harm the subsidizers and help the subsidizees. Yet nations often do subsidize their exports, overtly and covertly. Instead of welcoming cheap goods with open arms, the recipients of the subsidized merchandise usually complain, and sometimes slap on "anti-dumping" tariffs to keep cheap goods from being too cheap. Economists often tsk-tsk at all this, blaming both the subsidies and the tariffs on rent-seeking politically connected manufacturers. It's all "protectionism", they say.

A fair review of the history of "protectionism" would be much more mixed than the economic mainstream would like us to believe, with their stories of comparative advantage and expanding production possibility frontiers. (Thankfully, economists like Dani Rodrik and Paul Krugman weave more nuanced tales, but still "protectionism" rates somewhere just below coprophagia on the economic profession's list of distasteful things.) In some times and places, trade barriers have served to isolate and impoverish people. In other times and places, tariffs have protected infant industries that grew into powerhouses in countries (like the United States) that otherwise might have remained agricultural backwaters. That said, I think we should avoid tariffs, not in deference to economic pseudoscience, but because they are stultifying. Intercourse across borders is a per se good. A mixed-up, intermingling world is better than one made up of insulated national tribes. We should avoid tariffs not because of their adverse economic consequences, but despite their potential economic benefits.

But subsidy is a different story. Export subsidies do not diminish international commerce, they, um, subsidize it. From a libertarian perspective, there is a strong case against tariffs. Trade restrictions prevent free people across borders from interacting as they wish. But subsidies restrict no one. Sure, libertarians might complain of the wealth expropriated to fund the subsidy, but that critique applies to nearly all functions of modern government. Until we abolish public schools and the NIH, there's no reason we shouldn't have export subsidies.

The more serious case against subsidies is that they are "distortionary". But for even the most ham-handed sort of subsidies, where governments favor particular firms or industries, it is not at all clear that this is so. Investment is not a "distortion", even though it involves accepting an up-front cost. When local governments offer tax abatements, free infrastructure, and other perqs to attract economic activity, there's a clear payoff from taxpayers to particular private parties. Yet sometimes these inducements do pay for themselves, in financial terms as growth increases the long-term tax base by more than the upfront costs, and in nonfinancial terms as residents reap direct and indirect benefits from prosperity of place. Sure, governments make poor investments sometimes, whether corruptly or out of innocent miscalculation. Firm managers also make bad investments, and sometimes their motivations in doing so are not aligned with the welfare of shareholders. Sometimes firms are large, and capable of investing on a scale that deters potentially superior upstarts from entering a market. But we don't prohibit corporate investment as "distortionary". In both the public and private sector, restricting investment implies preventing potentially welfare-enhancing projects from taking root. Subsidies, when they are not a form of corruption, are a form of investment. We should be very careful in designing public subsidies to private parties, since the potential for crooked dealing is obvious. But forbidding subsidy outright is prima facie welfare destructive. Preventing governments from internalizing the external benefits their communities would receive from economic development would itself be "distortionary". (I dislike the language of optimality and distortion favored by economists. But when in Rome...)

The example of the United States is often held up as a model of a free-trade zone, as a reducto ad absurdiam. If protectionism is such a good idea, asks some supercilious hypothetical interlocutor, why shouldn't we have tariffs between Tennessee and Alabama? Of course we don't, and shouldn't. But Tennessee and Alabama can and do compete in bidding wars with firms deciding where they ought to put their factories. The "free trade" that has worked so well among the 50 United States is actually a trade regime involving ubiquitous and competitive subsidies. Maybe that's not a flaw, but a feature.

At this moment, there's a fear that "Smoot-Hawley", "beggar-thy-neighbor" protectionism will take hold, condemning us to a depression more harmful than the one we already face. If insufficient aggregate demand is the problem, then competitive tariffs are a negative sum game: They not only confine demand within borders, but they eliminate demand that would otherwise exist for goods and services that could be provided internationally. So, the fear of tariffs is not misplaced.

But competitive export subsidies are a different thing entirely. If the people from whom funds are borrowed or taxed would otherwise have saved, then export subsidies can increase effective aggregate demand. A trade war in which the nations of the world strive to outgive one another in order to help support their own industries would amount to a collaborative global stimulus. Angloamerican economists are tut-tutting over how "surplus countries" aren't doing their part in stimulating domestic consumption. Instead, export-heavy nations are stimulating consumption elsewhere, by stepping up their export subsidies. If China wants to support American consumption, then why shouldn't America support Chinese consumption? Rather than digging holes again and filling them back in again, why not give the world's "bottom billion" perishable gift cards redeemable for US goods and services, and let the jobs follow?

I don't think this is only a matter of "depression economics". It really is better to give than to receive, even in good times. But it is impolite to give but then refuse the gifts of others. And it is best to be up front about what you are doing. Making "loans" that are unlikely to be repaid is the worst form of giving. Everyone ends up unhappy when the inevitable comes to pass.

I started with the example of China, and I'll end with it. A year or two ago, China looked unstoppable, but suddenly conventional wisdom is that chickens are coming home to roost. China has subsidized exports, but its subsidy has been synthetic, implicit and deniable, and therein lies its problem. As Brad Setser has described for years, in order to maintain a "crawling" currency peg, China's central bank has been forced to purchase US dollar assets on which it must expect an eventual loss in real terms. China's subsidy to foreign consumers has been hidden in this overpayment. China's policy of giving worked very well for it, but executing that policy by pretending to lend rather than to give has put the nation in a bind. The technocrats responsible for China's huge currency reserves must continually expand their losses by purchasing more dollars to keep the value of the dollar high and hide the costs of subsidies already granted. If they do not, the exposure of large financial losses might create a firestorm of domestic outrage. China's central bank might be able to hide reserve losses by engineering a large domestic inflation, so that its US dollar portfolio does not lose value in nominal terms. In either case, even though the development gains were almost certainly worth the financial cost of China's export support, China's leaders face a problem since they pretended there was no subsidy when in fact the subsidy was very large.

It would have been better for China as well as for its trade partners (who face traumatic currency devaluations) if its policies had involved explicit, sustainable, and broad-based subsidies to foreign consumers. Explicit subsidies paid over time are more politically palatable than sharp losses suddenly revealed. China's covert, financial-engineered subsidies relied upon complex chains of financial intermediation, which eventually could not withstand the stress. China is still trying to subsidize, but lending to the US Treasury no longer translates to increased consumption by American consumers. China's approach to subsidy contributed to instability in the financial arrangements of its customers, which has unsurprisingly boomeranged, creating economic instability in China.

Here is my proposal for the WTO. I know it will be greeted enthusiastically. Explicit export subsidies in the form of time-limited direct-to-consumer vouchers redeemable towards substantially all of a country's domestically produced goods and services should be deemed permissible, and the inevitable bureaucracy should be created to quibble over the terms of the institutionalized subsidy. Nations may choose to opt out of the program, but if they wish to offer subsidies, they must accept all other nations' subsidies. (Nations may accept subsidies without offering them, though.) Each subsidizing nation then sets an annual lump-sum amount, which is distributed in the form of equal-valued vouchers to adults in all participating nations worldwide. (Goverments that cannot brook direct-to-consumer payments would be excluded both from offering or accepting subsidies under the program.) Vouchers would be transferrable, but redeemable only by non-residents of the issuing country, for delivery outside of the issuing country. (Yes, for electronically deliverable services that might be hard to enforce. But that's what we have bureaucracies for.) In particular, subsidy vouchers could be bought and sold on organized exchanges, so that recipients who need food more than imports could sell them, for example to entrepreneurs hoping to purchase foreign capital goods at a discount.

I know this will grate on some of my "free-trade" luvin' readers, but please compare this proposal with the actual status quo rather than hypothetical optimization problem. Governments will subsidize, sometime corruptly, sometimes mistakenly, and sometimes because it is a good idea that they do so. This scheme does not directly address narrowly tailored subsidies (e.g. US farm subsidies), but it does provide an alternative and "less distorting" means by which nations can broadly support their tradable industries while picking winners and losers as little as possible. It will also provide an alternative to the current practice of synthetic subsidy via currency and financial market intervention, which has led us to the brink of depression and dramatically increased the likelihood of serious conflict, economic or otherwise, between major powers. Since governments will always subsidize, we should try to devise and institutionalize least-harmful-means by which governments can do what they will (and sometimes should) do. This proposal avoids government picking of winners and losers, encouraging governments to subsidize tradables very broadly defined but let markets fill in the details. It prevents governments from targeting and undermining tradables production in particular countries. It avoids the obscenity of the current decade, wherein the mechanics by which export subsides were arranged meant that the wealth transfer went primarily towards the consumption of the already wealthy (owners of real estate or financial assets). The aggregate demand required to mobilize China might have been generated by entrepreneurs building factories in Africa rather than homeowners buying lawn furniture in America. Also, the structure of the proposed subsidy means that poor countries can choose to accept it as a form of foreign aid whose direct-to-consumer requirement might limit corrupt misuse, and whose breadth renders the subsidy less harmful to domestic producers than, say, dumping underpriced grains onto the market in the name of charity.

I think this is a pretty good idea. Tell me why I am wrong.