Source: Getty

China and the History of U.S. Growth Models

The Chinese development model is an old one and can trace its roots at least as far back as the infant industry protection, internal improvements, and system of national finance of the American System of the 1820s and 1830s. Understanding why the many precedents for its growth model have succeeded in some few cases and failed in others will help us enormously in understanding China’s prospects.

Published on February 28, 2017

Note: Because of rather poor management of my blog, some entries seem to have disappeared. Fortunately for me, several other sites reproduce many of my blog posts, so when I have had to look for them and can remember the title, I have usually been able to find them. From time to time, if I think they might still be useful, I will repost them here on the Carnegie site with a little bit of light editing to correct typos and add links. The first blog entry that will receive this treatment is a February 15, 2013, post, in which I tried to place the Chinese growth model in a historical context that begins with the famous American System developed in the early nineteenth century.

As regular readers know, I have often argued that the Chinese development model is an old one and can trace its roots at least as far back as the American System of the 1820s and 1830s. This system was itself based primarily on the works of the astonishing first U.S. Secretary of the Treasury, Alexander Hamilton. (See especially his three main reports to Congress: his “First Report on The Public Credit” of January 14, 1790; his “Second Report on The Public Credit” of December 13, 1790, and most importantly his brilliant “Report on Manufactures” of December 5, 1791. The “Second Report on The Public Credit” is sometimes also known as the “Report on a National Bank.”)

This development model was also implicitly part of the debate in France that led to one of the most important financial innovations of the nineteenth century, the creation of the Crédit Mobilier in France in 1852. The importance of this is discussed, for example, in Alexander Gershenkron’s essay “Economic Backwardness in Historical Perspective” (pages 11–16). The debate concerned one of the great economic questions in France, especially after the defeat of Napoleon: why had England, a country that one hundred years earlier had been poorer than France, managed to surpass France and all other countries economically and technologically, even though in the pure sciences and engineering the French were at least the equal to the British and perhaps superior?

One obvious reason had to do with the financing of the commercial application of new technology. The French banking system, dominated by rentiers and the landed aristocracy, seemed to specialize in protecting savers, in part by mobilizing capital and investing in gold or in government obligations. The English banking system did this too, but it also seemed much more willing to finance infrastructure and manufacturing capacity.

In fact, more generally I have argued that the main reason industrial revolutions have occurred largely in England and the United States is because industrial revolutions are not driven by scientific developments but rather by the commercial application of scientific developments. For this to happen, it seems that a robust financing system is key. England, and later the United States, benefitted from a financial system that seemed to do better than others in financing new infrastructure and technological ventures.

A well-functioning financial system, one that allocates capital to new ventures, in other words, may have been the key difference between England and France at the end of the eighteenth century, and for this some historians blame the brilliant but erratic John Law and his Mississippi Bubble. This concern about the inefficient French banking system led to the creation of Crédit Mobilier, whose role was to break the constraints of the existing Rothschild-dominated financial system; mobilize the savings of the middle classes; and allocate these savings toward financial projects, such as infrastructure development, that would, over the longer term lead to more rapid economic development.

I will come back to this issue of the financial system, but the point here is that there have been many versions of this development model. At least two major economic theoreticians—the German Friedrich List in the nineteenth century and the Ukrainian-American Alexander Gershenkron in the twentieth—have formally described variations on the investment-driven growth model. Michael Hudson, one of my favorite economic thinkers, wrote a brilliant and provocative book (Trade, Development and Foreign Debt) twenty years ago; it traces many aspects of this model to debates in England at the end of the eighteenth century.

Aside from Alexander Hamilton, its intellectual and political godfather, the main proponents of the American System were figures like Henry Clay, Henry and Matthew Cary, John Calhoun, and even Abraham Lincoln himself. Their vision of economic policymaking was looked down upon as naïve and even foolish by most American academic economists— schooled as they were in the laissez faire doctrines then fashionable in England. But I think it is hard for any economic historian not to feel relieved that neither the academics nor the Jeffersonian and Jacksonian factions had the clout to force what they deemed good economic policy onto U.S. development. America got rich in part by doing the wrong things.

Many countries in which the academics had real influence at the time—Chile in the 1860s under the tutelage of the famous French economist Jean Gustave Courcelle-Seneuil, for example, or Mexico at the turn of the century under the expert guidance of José Y. Limantour, finance minister under then president Porfirio Díaz—never achieved the kind of growth that the less capable student-countries experienced. I write about some of these cases in my 1996 Foreign Affairs article, for anyone who might be interested.

To get back to the main story, in another book (America’s Protectionist Take-off, 1815-1914), Michael Hudson refers to a leading member of the second generation of proponents of the American System, E. Pechine Smith. What is especially interesting about Smith in the context of China is that in 1872 he was invited to Japan to serve as adviser to the Mikado, becoming the first of a stream of economists and lawyers—most of them proponents of the American System—to advise and help shape Japanese development after the Meiji Restoration.

Smith thus creates a direct link between the American System and the Chinese development model. It was, of course, the post-war Japanese development model, itself based on Japan’s experience of economic development during and after the Meiji Restoration, that became the standard for policymakers throughout East Asia and China. I think of China’s growth model as merely a more muscular version of the Japanese or East Asian growth model, which is itself partly based on the American experience.

There were three key elements of the American System. Historian Michael Lind, in one of his economic histories of the United States, described them as:

  • infant industry tariffs,
  • internal improvements, and
  • a sound system of national finance

These three elements are at the heart, explicitly or implicitly, of every variation of the investment-led development model adopted by number of countries in the last century— including Germany in the 1930s, the Soviet Union in the early Cold War period, Brazil during the Brazilian miracle, South Korea after the Korean War, Japan before 1990, and China today—to name just the most important and obvious cases. For this reason, I think it makes sense to discuss each of these three elements in a little more detail.

Infant Industry Tariffs

The infant industry argument is fairly well known. I believe Alexander Hamilton was the first person to use the phrase, and the reasoning behind his thinking was straightforward. American manufacturing could not compete with the far superior British, and according to then- (and now) fashionable economic theories based on Adam Smith and David Ricardo, the implications for trade policy were obvious. Americans should specialize in areas where they were economically superior to the British—agriculture, for the most part—and economic policy should consist of converting U.S. agriculture to the production of cash crops—tobacco, rice, sugar, wheat, and, most importantly, cotton—maximizing that production and exchanging these goods for cheaper and superior British manufactured items.

In this way, as Ricardo brilliantly proved, and assuming a static distribution of comparative advantages, with each country specializing in its comparative advantage, global production would be maximized and through trade both the British and the Americans would be better off. While most academic U.S. economists and the commodity-producing South embraced free trade, Alexander Hamilton and his followers, mainly in the northeast, did not. (In fact, differing views over free trade as well as over slavery and state rights were at the heart of the North-South conflict that led eventually to the Civil War.)

Hamilton was convinced that it was important for the United States to develop its own manufacturing base because, as he explained in his 1791 congressional report, he believed that productivity growth was likely to be much higher in manufacturing than in agriculture or mineral extraction. Contrary to David Ricardo, in other words, Hamilton believed that comparative advantage was not static and could be forced to change in ways that benefitted less productive countries. What is more, he thought manufacturing could employ a greater variety of people and was not subject to seasonal fluctuations or fluctuation in access to minerals.

Given much higher British efficiency and productivity, which translated into much lower prices even with higher transportation costs, how could Americans compete? They could do it the same way the British had done to compete with the superior Dutch a century earlier. The United States had to impose tariffs and other measures to raise the cost of foreign manufacturers sufficiently to allow their American counterparts to undersell them in the U.S. market. In addition, Americans had to acquire as much British technological expertise and capacity as possible (which usually happened, I should add, in the form of intellectual property theft).

This the United States did. In fact, I believe every country that has managed the transition from underdeveloped to developed-country status—including Germany, Japan, and South Korea—has done it behind high explicit or implicit trade tariffs and stolen intellectual property. (This holds true with, perhaps, the exception of one or two trading entrepôts like Singapore and Hong Kong, although even that is debatable.) The idea that countries get rich under conditions of free trade has very little historical support, and it is far more likely that rich countries discover the benefits of free trade only after they get rich, while poor countries that embrace free trade too eagerly (think of Colombia and Chile in the late nineteenth century, which were stellar students of economic orthodoxy) almost never get rich. This has been the case unless, like Haiti in the eighteenth century or Kuwait today, they are massive exporters of a very valuable commodity (sugar, in the case of Haiti, which was the richest country in the world per capita during a good part of the late eighteenth century).

But rather than just embrace protection, I would add that there is one very important caveat. Many countries have protected their infant industries, and often for many decades, and yet very few have made the transition to developed-country status. Understanding why protection seems to work in some cases and not in others might have very important implications for China. I won’t pretend to have answered this question fully, but I suspect the difference between the countries that saw such rapid productivity growth behind infant industry protection that they were eventually able to compete on their own, and those that didn’t, may have had to do with the structure of domestic competition.

Specifically, it is not enough to protect industry from foreign competition. There must be a spur to domestic innovation, and this spur is probably competition that leads to advances in productivity and management organization. I would argue, for example, that countries that protected domestic industry but allowed their domestic markets to be captured and dominated by national champions were never likely to develop in the way the United States did in the nineteenth century.

I would also argue that companies that receive substantial subsidies from the state also fail to develop in the necessary way because rather than force management to improve economic efficiency as a way of overcoming their domestic rivals, these countries encourage managers to compete by trying to gain greater access to those subsidies. Why innovate when it is far more profitable to demand greater subsidies, especially when subsidized companies can easily put innovative companies out of business?

Last April, for example, I wrote about plans by Wuhan Iron & Steel, China’s fourth-largest steel producer, to invest $4.7 billion in the pork production industry. The company’s management argued that they could compete with traditional agro-businesses not because steel makers were somehow more efficient than farmers, but rather because their size and clout made it easier for them to get cheap capital and to get government approvals. They were able to invest in an industry they knew little about, in other words, because they knew they could extract economic rent. This clearly is not a good use of protection.

The lessons for China, if I am right, are that China should forego the idea of nurturing national champions and should instead encourage brutal domestic competition. Beijing should also eliminate subsidies to production, the most important being cheap and unlimited credit, because senior managers of Chinese companies rationally spend more time on increasing access to these subsidies than on innovation, an area in which, in spite of the almost absurd hype of recent years, China fares very, very poorly.

There is nothing wrong with protecting domestic industry, but the point is to create an incentive structure that forces increasing efficiency behind barriers of protection. The difficulty, of course, is that trade barriers and other forms of subsidy and protection can become highly addictive, and the beneficiaries, especially if they are national champions, can become politically very powerful. In that case, they are likely to work actively both to maintain protection and to limit efficiency-enhancing domestic competition. It was Friedrich Engels, not often seen as a champion of capitalist competition, writing to Edward Bernstein in 1881, who said that “the worst of protection is that when you once have got it, you cannot easily get rid of it.”

Internal Improvements

The second element of the American System was internal improvement, which today we would probably call infrastructure spending. Proponents of the American System demanded that the national and state governments design; finance; and construct canals, bridges, ports, railroads, toll roads, and a wide variety of communication and transportation facilities that would allow businesses to operate more efficiently and profitably. In some cases, these projects were paid for directly (tolls, for example), and in other cases they were paid for tax revenues generated by higher levels of economic activity.

It is easy to make a case for state involvement in infrastructure investment. The costs of infrastructure can be very high, while even if the benefits are much higher, they are likely to be diffused throughout the economy, making it hard for any individual company to justify absorbing the costs of investment. In this case, the state should fund infrastructure investment and pay for it through the higher taxes generated by greater economic activity.

For me, the interesting question, especially in the Chinese context, is not whether the state should build infrastructure but rather how much it should build. In fact, this is one of the greatest sources of confusion in the whole China debate. Most China bulls implicitly assume that infrastructure spending is always good and that the optimal amount of infrastructure is more or less the same for every country. This is what allows them to compare China’s per capita capital stock with that of the United States and Japan and conclude that China still has a huge amount of investing to do because its capital stock per capita is so much lower.

But this is completely wrong and even nonsensical. Infrastructure investment is like any other investment in that it is only economically justified if the total economic value created by the investment exceeds the total economic cost associated with that investment. If a country spends more on infrastructure than the resulting increase in productivity, more infrastructure makes it poorer, not richer.

In China, we have problems with both sides of the equation. First, we don’t know what the true economic cost of investment in China might be. To calculate the true cost, we need to add not just the direct costs but also all the implicit and explicit subsidies, most of which are hidden or hard to calculate.

The most important of these subsidies tends to be the interest rate subsidy, and this can be substantial. If interest rates in China are set artificially low by 5 percentage points, for example (which is a reasonable estimate), an investment of $100 million receives an additional subsidy of $5 million for every year that the loan funding the investment is outstanding—and loans are almost never repaid in China. Over ten to twenty years of outstanding debt, this can add 30­–40 percent to the initial cost of the investment. This means that the recognized cost of an infrastructure project is much lower than the true economic cost, with the difference being buried in explicit and implicit subsidies.

But the bigger problem is in the value created by the investment. We can think of the value of infrastructure primarily as a function of the value of labor saved. In countries with very low levels of productivity, each hour of labor saved is less valuable than each hour saved in countries with high levels of productivity. For this reason, less productive countries should have much lower capital stock per capita than more productive countries.

This should be obvious, but it seems that often it isn’t. When analysts point to high-quality infrastructure in China whose quality exceeds comparable infrastructure in rich countries, this is not necessarily a good thing. It might just be an example of the amount of waste you can achieve when spending is heavily subsidized, when there are strong political (or pecuniary) incentives for expanding investment, and when there is limited transparency and accountability.

Other things matter too. If a country has low levels of social capital—if it is hard to set up a business, if less efficient businesses with government connections can successfully compete with more efficient businesses without government connections, if the legal and political structure creates problems in corporate governance (the agency problem, especially), if the legal framework is weak, if property rights are not respected, if intellectual property can easily be lost—then much infrastructure spending is likely to be wasted.

In fact, it turns out that it may be far more efficient to focus on improving, say, the legal framework than to build more airports, even though (and perhaps because) building airports generates more growth (and wealth for the politically connected) today. Weak social capital becomes a constraint on the ability to extract value from infrastructure, and this constraint is very high in poor countries with weak institutional frameworks.

Journey to the West

This issue of how much investment is enough is a very important topic that deserves much more discussion, but I think there is a very good example of why we need to be worried about how useful additional infrastructure investment in China might be. This shows up most clearly in China’s push to create development in the western part of the country.

Often when I question the economic value of China’s push to the western, poorer parts of the country (by the way economic value is not the same as social or political value, the latter of which may nonetheless justify projects that are not economically viable), I am almost always treated with the story of the American West. In the nineteenth century, as everyone knows, the United States went west, and most economists agree that this made economic sense for the country and was an important part of the process that led it to becoming the wealthiest and most productive country in history.

But we must be very careful about drawing lessons from the American experience. The United States is not the only country in history that went west. Several other countries did so too, but for some reason we ignore their experiences altogether when we discuss China. Brazil, for example, went west and north in the 1950s and 1960s, as it expanded from the rich southern coastal areas into the Amazon and the Caribbean. The Soviet Union did something similar after the Second World War as it went east into Siberia.

Most economists today agree that the Brazilian and Soviet experiences were economically unsuccessful and left those countries burdened with such enormous debts that they were at least partly to blame for Brazil’s debt crisis in the 1980s and the collapse of the Soviet economy in the 1970s. It turns out, in other words, that there are both successful and unsuccessful precedents for China’s going west.

What are the differences and how do they apply to China? Again, I can’t say that I can fully understand or explain them, but one major difference leaps out. In the United States, it was private individuals, seeking profitable opportunities, that led the move into the American West, and government investment followed. In Brazil and the Soviet Union, however, there was little incentive for private individuals to lead the process. It was the government that led, and private businesses followed only because government spending created great opportunities for profit. Once government spending stopped, so did business.

My very preliminary conclusion is that large-scale government ambitions allied to strong political motivation and funded by cheap and easy access to credit can lead very easily to the wrong kinds of investment programs. The U.S. experiences of government investment in the nineteenth century, in other words, may be a very poor precedent for understanding China’s current policy of increasing investment spending, especially in the poor western part of the country.

Brazil and the Soviet Union may be much better precedents. At the very least, these gloomier experiences should not be ignored when we think of China’s policies. Going west isn’t always a great idea from an economic point of view and has led to at least as many, and probably more, bad outcomes as good outcomes. It is not clear why these lessons cannot possibly be applied to China.

A Sound System of National Finance

The third pillar of the American System was the creation of an appropriate financial system. But what does that mean? It is hard to describe the American financial system in the nineteenth century as stable and well-functioning. In fact, the American banking system was chaotic, prone to crises, mismanaged, and often fraudulent; yet the United States grew very rapidly during that time.

China’s banking system, on the other hand, is far more stable—in fact, the favorite cliché of Chinese bankers is that while the system may not be efficient, it is very stable. What makes the Chinese banking system stable, of course, is that it is widely believed that the government stands fully behind the banks. It makes no difference, in other words, how weak a credit allocation decision is, because by controlling credit and the deposit rate, and by limiting alternatives for Chinese savers, the government guarantees both the liquidity and solvency of the banking system. As long as government credibility is intact, the banking system is unlikely to fail.

In that sense, you can easily make the case the Chinese banks today are sounder than American banks in the nineteenth century. This might bode well for the future of the financial system in the short term, but in the long term it is not clear to me that monetary soundness and financial stability are necessarily correlated with more rapid growth.

I say this because I have seen no evidence that countries with sound and conservative financial systems grow faster than countries with looser and riskier financial systems (although they do seem to have fewer financial crises). In fact, I have more than once made reference to Belgian bank historian Raymond de Roover’s provocative and profound comment that “perhaps one could say that reckless banking, while causing many losses to creditors, speeded up the economic development of the United States, while sound banking may have retarded the economic development of Canada.” Canada was blessed (or cursed, according to de Roover) in the nineteenth century with being part of Britain, and so inheriting England’s much better-managed financial system.

Reckless banking is hard to define. Certainly, it is easy to make the case that Chinese banking has been reckless, especially in recent years, but it is a very different type of recklessness. Once again I cannot say with complete confidence how China’s version of its development model differs meaningfully with the American System on the subject of banking, but I would suggest there are at least two very important differences.

First, the American financial system then (and now) has been very good at providing money to risky new ventures. It provides capital on the basis not only of asset value but, more importantly, on future growth expectations, and risk-taking has been actively rewarded. In China, it isn’t clear that this is the case at all. Chinese banks favor large, well-connected, and often inefficient giants at the expense of risk-takers.

Second, although both systems were prone to bad lending, the American banking system tended to correct very quickly—in the form of a crisis—and bad loans were written down and liquidated almost immediately. This was certainly painful in the short term—especially if you were a depositor in the affected bank—but by writing down loans and liquidating assets, three important objectives were achieved. Financial distress costs were quickly eliminated (writing down debt does that in ways I won’t get into because they are well-known and much discussed in corporate finance theory); capital allocation was driven by profitability, not by implicit guarantees; and assets were returned to economic usefulness quickly.

A classic example of the last of these objectives may be the response to the railway bubble of the 1860s. During and after the 1873 crisis, a number of railroads went bankrupt, including major lines like the Union Pacific and the Northern Pacific, the latter of which even brought down Jay Cooke & Company, the leading financier of the U.S. government during the Civil War. After the crisis, some major railway bonds traded as low as 15–20 percent of their original face value, and so they were purchased and reorganized at huge discounts. The new buyers were consequently able to cut freight and passenger costs dramatically, in some cases by over 50 percent, while still earning more than enough to cover the costs of buying the railroads. This led to a collapse in transportation costs in the United States.

Liquidation, in other words, provides an important economic value to the economy. It allows assets to be re-priced, which creates a boost to the economy and prevents those assets from acting as a deadweight loss. If the railroads hadn’t been liquidated, in other words, any reduction in costs was likely to be minimal and the railroads would have been far less useful to the development of the U.S. economy.

Comparing Development Models

This blog entry is long, and I plan to write about this a lot more in the future, but for now I think it makes sense to summarize some of the important points about the American System and other similar growth models, like the Chinese version.

  1. Infant industry protection has worked to promote long-term development under certain conditions and has not worked under other conditions. I would argue that the key difference is that in the former case there were powerful forces that drove managerial and technological innovation and rapid growth in efficiency.

    In the U.S. case, this seems to have been brutal domestic competition. If China wants to benefit from its own protection of infant industry, it is important that there be similar domestic drivers of innovation and efficiency. Note that access to cheap capital cannot be such a driver, even though it is one of the main sources of Chinese competitiveness. Access to cheap capital is just another way to protect infant industries from foreign competition.
  2. Every country that has become sustainably rich has had significant government investment in infrastructure, but not every country that has had significant government investment in infrastructure has become sustainably rich. On the contrary, there are many cases of countries with extraordinarily high levels of infrastructure investment that have grown for a period and then faltered.

    I would argue that the difference is almost certainly the extent of capital misallocation. In some countries, it has been much easier for policymakers to drive capital expenditures, and in those countries, it seems to have been relatively easy to waste investment. If this is the case in China, as I believe it is, the key issue for China is to rein in its spending and develop an alternative and better way to allocate capital.

    The point is that there is a natural limit to infrastructure spending, and this limit is often imposed by institutional distortions in the market economy. When this natural limit is reached, more investment in infrastructure can be wealth-destroying, not wealth-enhancing, in which case it is far better to cut back on investment and to focus on reducing the institutional constraints to more productive use of capital, such as weak corporate governance and a weak legal framework. The pace of infrastructure investment cannot exceed the pace of institutional reform for very long without itself becoming a problem.
  3. Any economy looking to achieve sustainable long-term growth must have what can be deemed a good financial system that allocates capital efficiently and rewards the correct level of risk-taking. It is hard to determine what the characteristics of a good financial system are, but we shouldn’t be too quick to assume that this has to do with stability.

    What’s more, while obviously the capital allocation process is vitally important, I would also suggest that the liquidation of bad loans is just as important. Bad loans, as Japan showed us in the past two decades, can become a serious impediment to growth, in part because financial distress distorts management incentives in the way widely understood and described in corporate finance theory, and in part because they retard the process by which bad investment is absorbed by the economy.
  4. One thing I have not discussed above is the role of wages. The American System was developed in opposition to the then-dominant economic theories of Adam Smith and David Ricardo, in part because classic British economic theory seemed to imply that reductions in wages were positive for economic growth by making manufacturing more competitive in international markets. A main focus of the American System, however, was to explain what policies the United States, with its much higher wages than in Europe at the time, had to engineer to generate rapid growth. Sustaining high wages, in fact, became one of the key aspects of the American System.

    The Japanese version of this development model, as well as many of the various versions implemented in other countries throughout the twentieth century, shared its view of wages not with the American System, but rather with classic British economic theory. Rather than take steps to force up wages and keep them high—thereby both driving productivity growth and creating a large domestic consumption market for American producers—many of the later versions of the American System sought to repress growth in household income relative to total production as a way of improving international competitiveness. This is perhaps the main reason why the United States, unlike many other countries that have implemented similar development strategies in the twentieth Century, tended to run large current account deficits for much of the nineteenth Century.

    This different focus on whether high wages are to be encouraged or discouraged, I believe—although very little discussed in the theoretical literature as far as I know—nonetheless is perhaps the most important difference between the American development model and its many descendants in the twentieth and twenty-first centuries. I would even argue, although I cannot prove it, that one consequence of this difference is that growth in demand tends to be more sustainable when it is balanced between growth in both consumption and investment.

In analyzing China’s growth in the past three decades, we seem to forget that there have been many so-called growth miracles in the past two hundred years. Some have been sustainable and have led to developed-country status, but many, if not most, were ultimately unsustainable. Nearly all of the various versions have had some similar characteristics—most obviously infant industry protection, state-led investment in infrastructure, and a financial system that disproportionately favored producers at the expense of savers. But the way these characteristics played out were very different, in large part because the institutional structure of the economy and the financial sector created a very different set of incentives.

I would argue that, in understanding China’s growth and its sustainability, we need to have a clear understanding of why these characteristics worked in some cases and not in others. Most economists who focus on China seem to know little about economic history, and when they do, their knowledge tends to be limited to a very superficial understanding of U.S. economic history. But there are many precedents for what is happening in China and not all suggest that further Chinese growth is inevitable.

On the contrary, the historical precedents should worry us. In most cases, they suggest that China has a very difficult adjustment ahead of it, and the closest parallels to its decades of miracle growth suggest unfavorable outcomes. Understanding why the growth model has succeeded in some few cases and failed in most will help us enormously in understanding China’s prospects.