international analysis and commentary

The one percent myth and the IMF-Piketty mistake


Capitalism is not a rigid orthodoxy. If we look at the long-term evidence, two facts stand out. First, different versions of free market capitalism in North America, Western Europe, and Australia hold the unmatched record for growth of living standards and personal freedom. Tiny colonies on the eastern edge of the North American continent became the world’s most successful economy. Without capitalism, the welfare state could not have redistributed as much as it has in Western Europe and North America.

We should recognize that capitalism is a human institution. As such, it is imperfect. Some will steal, cheat, lie, engage in fraud. We do not endorse or condone such behavior, but we recognize that these are human failings that do not disappear or lessen when power shifts from markets and capitalists to politicians. Most often the opposite is true. As Lord Acton warned: “Power corrupts and absolute power corrupts absolutely.” Medieval Christianity recognized that mankind was not perfect. Human institutions, including markets, require rules and enforcement to achieve their promise.

The one percent
One perennial complaint of welfare state advocates and the most frequent complaint about capitalism is that it benefits the rich at the expense of the middle and lower classes. This claim has been revived recently in several countries. The proponents call for increased taxation of the highest incomes to finance increased redistribution to the lowest income groups including those with no earned income. Some claim that the spending increase generated in this way will increase economic growth.

Support for the alleged social benefits of setting much higher marginal tax rates for the highest incomes has now been endorsed by the International Monetary Fund, based heavily on research by two French economists named Thomas Piketty and Emanuel Saez. The two worked together on the faculty at MIT, where the current research director of the IMF was a professor. Like Piketty and Saez, he is also French. That alone may explain the preference for destructive policies of income redistribution because that is what France has done for many years.

Professor Piketty collected data on income distribution from approximately 20 countries over periods of different length. He concluded that raising the tax rate to 60% on the highest incomes and redistributing the receipts to the poor would increase spending and economic growth. The New York Times declared his book, Capital in the 21st Century one of the great achievements of modern economics. It put it in a class with Karl Marx’s Capital and John Maynard Keynes’s General Theory.

This lavish praise seems both wrong and extreme. I agree that in the past there was a notable negative association between economic growth and the spread between the shares of income going to the top 1, 5, or 10% of the earners and the share going to the remainder. The mistake is to conclude that narrowing the distribution contributes to growth. The far more plausible explanation is that economic growth in capitalist countries over the past two centuries contributed to a steep decline in the share of the top earners.

The crux of Piketty’s argument is that when economies grow the rate of return to capital exceeds the growth rate of output; so the share of earned income received by owners of capital rises relative to other sources of income. Since ownership of capital is concentrated in the wealthiest group, the distribution of income shifts in their favor.

This is an empirical statement, so it can be shown to be true or false. My reading of the history of economic development shows that it has typically been falsified. Growth and the highest share of the income distribution, the share of the 1%, fell relative to the 99% whenever we have long series on that share.

Simply put, Piketty, President Obama, and the IMF have the causality running the wrong way.  Taxing the rich to redistribute did not produce growth. On the contrary, capitalist growth reduced the share earned by the highest earners.

In my book Why Capitalism? (Oxford, 2012) I used some data on the share received by the top 1% that two Swedish economists gathered earlier. Piketty uses some of the same data. The seven countries in the Swedish data are the United States, the United Kingdom, Sweden, Australia, France, Canada, and the Netherlands. Of course, data on income distribution is never precise, but broad trends can be informative.

The data show a remarkable degree of uniformity.[1] The share of income received by the top 1% declined persistently from about 1910 to 1980. The share fell from an initial 20 to 25% to about 5 to 8%. Then the share rose in several of the countries, notably the US, the UK, Canada and Sweden. By the end of the sample, about 2005, the share in several of these countries is back to about 10%. The share of the top 1% of income earners in the US reaches 15%, more than half way back to where it was early in the 20th century. It is this rise that initiated the loud outcries about failure of modern capitalism to benefit the middle class.

It is impossible for Piketty, Saez or anyone else to show that the decline in all seven countries resulted from higher taxes on the highest incomes and redistribution to the poor. The main reason is that the welfare state did not exist in several of the countries and was relatively small in the others. In the United States, federal government spending was rarely more than 3 or 4% of total spending in non-war years until after 1930. Old age pensions didn’t start until the late 1930s, and healthcare spending did not expand until the late 1960s.

The timing in other countries differed. Sweden undertook welfare state spending in the 1930s, but substantial growth waited for the end of World War II, after economic growth resumed and incomes rose. The same is true for most other countries. Higher tax rates and increased social spending rose long after the share of the top 1% had declined from the early peak to the trough in 1980. In 1980, the share of the 1% ranged from 5% in Sweden with its larger welfare state to 8% in the United States.

Three major forces explain most of the decline. At the start of the series, private capital receives a relative high return because it is relatively scarce. As investment and the capital stock rose during the 20th century, the share of total income going to capital declined. The total income from capital increased but more slowly than total income. Since the highest income earners receive a disproportionate share of their income as earning on capital, their share fell. This is contrary to the key assumption that Piketty makes.

That alone does not explain the steep decline in the share received by the top 1%. During the early 20th century, the United States absorbed millions of immigrants, many unskilled. Many began employment at low wage jobs. Minimum wage laws did not come until the 1930s long after income shares narrowed. By working, the immigrants learned new skills; their productivity increased and with it their wages. That narrowed the gap between the incomes of the top and the bottom earners. But many did something else. They sent their children to colleges and universities where the children learned professional skills that earned middle class incomes.

This process continued in recent decades for immigrants from Korea, China, Mexico and Latin America. That history sends an important message. The growth of the middle class and the narrowing of the income distribution was in large part a result of working to acquire new skills and higher productivity.

Redistribution works against this process. It doesn’t reward work. It gives the unemployed and underemployed food stamps, healthcare, housing allowances, and even income. Instead of working, many learn to live on the government benefits supplementing them occasionally by working in the underground economy. Instead of acquiring productive skills, they learn how to live without working at regular jobs. That’s one way that the welfare state worked to increase the share of the highest paid 1% after 1980. The welfare state contributes also by weakening and even destroying family structure. Single family women are often on the bottom rung of the income distribution. Piketty does not include transfer payments in his measure of income and he does not reduce income of the 1% for taxes paid so his comparison is faulty and misleading.

A different process is at work now. The capital that is most highly rewarded is now human capital – the education and skill that produces innovations like the internet, the social media, popular computer apps, fracking and three dimensional manufacturing. The top 1% of the earners in the United States in any year includes people like Steve Jobs and Bill Gates who made the internet into a commercially successful, widely used means of communicating. But the top 1% everywhere also includes leading sports stars with unique skills and rock musicians with enormous popular appeal. The top 1% is a changing mix of people.

And what does Bill Gates do with his wealth? He spends much of it improving healthcare in Africa and improving education in the United States. An example is the effort to provide bed nets to protect Africans from malaria, an outcome that the public bureaucrats failed to achieve over decades.

The noisy political clamor about the rise in the share going to the top 1% gives no attention to the importance of education, skill, and an environment that encourages innovation. It is not an accident that most new products and much new music originate in the United States. Countries like Canada, Sweden, and the UK contribute also, welcoming foreign innovation and willing to reward those who bring new ideas to market successfully.

A strictly temporary recent influence on the income and wealth of those in the top 1 and 5% results from recent monetary policy. Expansive monetary policy begins by raising asset prices.  Since ownership of assets provide a disproportionate share of the income of the top 1% of all earners, their income rises relatively and absolutely. One of the peculiarities of the discussion of income shares is that many of those who advocate highly expansive monetary policy also complain actively about its effect on income distribution without recognizing the conflict in this idea.

The process that worked in the United States for several centuries worked well for many other countries in the past fifty years. Hong Kong is a remarkable example. Once China adopted capitalist methods, it moved millions out of low productivity agriculture, taught them new skills, and raised their wages. Korea sends its young to learn new skills in the United States. That enabled it to move successfully from a low wage provider of unskilled labor to a technically advanced country with a skilled labor force. And it increased freedom along with wealth. Other examples are Singapore, and Taiwan, relatively free, capitalist democracies. Add Chile, Peru and Mexico among others.

France is at the opposite pole. Mired in the hatred of capitalism and demands for redistribution, it has maintained its low share of income going to the top 1%. But the cost is high. The young and innovative leave France for the UK and other shores. The French get massive redistribution and sustained high unemployment for many who remain behind.

I  used some of the data from Piketty’s website to compare the income of the top 1% to the average growth rate for 2001-05 (before the recession) using countries with relatively high and low percentages earned or renewed by the top 1%.

The data show no relation between the income share and the five year average growth rate. In Singapore, the United Kingdom and Canada, the top 1% receives 12 to 14% of earned income.  Five year average growth (2001-05) rates are 3.7, 0.6, and 1.6 respectively.[2] Three countries with a relatively low percentage share for the 1% are Netherlands, about 6%, Sweden about 7, and France about 8. Five-year growth rates for the three countries are 0.8, 2.3, and 0.9.

The comparison suggests little association between the two series. Redistribution is only one factor affecting growth. The type of tax rate and other incentives including regulation and culture matter.

The lesson I take from the data is an old one, but one that is typically denied. Active market capitalism does more to narrow the distribution than active welfare state policies. It achieves that outcome by spreading opportunity and by increasing incentives to develop marketable skills whereas the welfare state discourages effort and encourages some young people to emigrate and others to leave the official labor force, accept the dole and augment incomes in unorganized markets. We see that strongly at work in France and growing rapidly in the United States.

Conclusion: What kind of future
The message that I draw from these comparisons is that one can augment Friedrich Hayek’s claim. Proponents of freedom and growth offer a more effective means of spreading the benefits of economic progress throughout society. And we accomplish these ends by treating people as individuals not as members of undifferentiated groups. Unlike the proponents of a larger government we recognize individual differences instead of treating people according to their race, sex, or nationality.  

We, who favor growth, freedom and personal responsibility, have as a principal task helping our fellow citizens to choose between a future that sustains growth, freedom, and personal responsibility or a more intrusive social welfare state. We can overcome pressures for more redistribution, more centralized regulation and less individual freedom by exploring the consequences and explaining them to our fellow citizens. Several studies by the IMF and others imply that the enormous debt and unfunded liability will reduce the US growth rate by as much as 20%. That’s one of many costs of its policies.

Getting voters to limit the welfare state will not be achieved easily. Working for us are two powerful facts. Most people prefer freedom to the regimentation of the social welfare state. And voters care about the future they leave to their children and grandchildren. The welfare state reduces growth and limits freedom. As Friedrich Hayek recognized so often, we offer higher growth, better living standards and greater freedom. Our challenge is to get others to recognize the consequences of their political choices.

This text is a revised version of the Presidential Address to the Mont Pelerin Society delivered in Hong Kong in September 2014

[1] Roine, Joseph and Waldenstrom, Daniel (2006). “The Evaluation of Top Incomes in an Egalitarian Society, Sweden 1903-2004.” Stockholm: Stockholm School of Economics.

[2] Five year growth rates are from Share of the 1% is from