Experimental feature

Listen to this article

Experimental feature

Jack Drake understands better than most Americans how strongly the US economy has performed over recent years. His job with a media company in Atlanta involves transcribing conference calls hosted by public companies to deliver financial information to analysts and investors.

“Almost every day, I listen to chief executives explaining how well their companies are doing,” he says.

But Mr Drake, 42, complains that the soaring corporate profits and robust economic growth he helps document are not reflected in his own financial circumstances. His $47,000 (£24,650, €36,850) annual salary has barely moved for five years. “Healthcare costs are up. Energy is up. But my income is standing still.”

Mr Drake is among millions of educated middle-class Americans seeing their pay stagnate and blaming that on technology and globalisation. “It would be hard to outsource my job because there is so much specialist knowledge and business jargon involved,” he says. “But it is used as an unspoken threat to keep wages down.”

Mr Drake plans to vote Democrat in next week’s congressional election. But he has little faith in either party to protect the economic interests of ordinary Americans. “I want the Republicans out but I don’t see the Democrats coming up with any good ideas,” he says.

The experience of people like Mr Drake helps explain why the Republican party is not getting the political credit that policymakers in the administration of President George W. Bush think it deserves for the state of the US economy. At a time when the party is already under siege over Iraq, failure to win credit for economic growth could prove costly.

The nub of the problem for the Republicans is this: since 2000 there has been a striking disparity between growth in productivity and gross domestic product on the one hand and growth in wages for the average American worker on the other. Economists call this phenomenon median wage stagnation.

The gap between productivity and wages

Median measures give the best picture of what is happening to the middle class because, unlike mean or average wages, median wages are not pulled upwards by rapid gains at the top. As the joke goes: Bill Gates walks into a bar and, on average, everyone there becomes a millionaire. But the median does not change.

Between 2000 and 2005, the US economy grew by 12 per cent in real terms and productivity, measured by output per hour worked in the business sector, rose 17 per cent. Over the same period, the median hourly wage – the wage the average American takes home – rose only 3 per cent in real (inflation-adjusted) terms. That compares with a 12 per cent gain in the previous five years. Real median family income fell every year from 2000 to 2004. It increased last year but is still lower than it was in 2000.

Jared Bernstein, an official in Bill Clinton’s administration and now at the Economic Policy Institute, says the gap between productivity and median wage growth “is the most significant economic challenge of the day”. He adds: “Workers have a right to be concerned about the large and growing gap between productivity growth and their pay cheques. They are working harder and smarter, baking a bigger pie more efficiently, but ending up with smaller slices.”

Median wage stagnation is unlikely to be the only reason the Republicans are not getting credit for strong economic growth – angst over Iraq is almost certainly poisoning voters’ views on a wide range of unrelated subjects. But it is fair to say that the Republicans would be in a stronger position if the average worker’s pay had better kept pace with overall economic growth since Mr Bush came to power.

In his first speech as Treasury secretary, Hank Paulson admitted that median wage stagnation was a problem, saying that “amid this country’s strong economic expansion, many Americans simply are not feeling the benefits”. Ben Bernanke, chairman of the Federal Reserve, has also aired worries about wage and income trends. He told senators in July that “inequality is potentially a concern for the US economy…to the extent that incomes and wealth are spreading apart, I think that is not a good trend.”

But while the basic facts are undeniable, there is heated debate as to whether there is anything unusual about the recent period of median wage stagnation – and whether it is explained by different policies pursued by Democrat and Republican administrations or by economic trends and cycles that politicians can do little to correct.

Wage inequality

Gene Sperling, a former chairman of the Clinton council of economic advisers who is now at the Council for Foreign Relations, draws a pointed contrast between the Bush years and the later Clinton years. “From 1995 to 2000, as America became more productive, typical workers, even low-income workers, were sharing in the wage gains,” he says. “Over the last five years, there has been the continuation of productivity growth but with a historically small share of the gains going to typical workers.”

Republicans see this comparison as fundamentally unfair. Ed Lazear, current chairman of the Bush council of economic advisers, says it is common for productivity and real wages to diverge in the first half of a business cycle. “Wage growth sometimes lags productivity growth, especially coming out of recessions,” he says. Then, as the cycle matures, “real wages begin to catch up with productivity”.

What is different this time, he says, is that unforeseen increases in the price of oil from 2003 onwards robbed workers of what would otherwise have been decent real wage gains, based on the expected rate of inflation.

There is undoubtedly some truth in this but it is not the whole story. To see why real wages for the average worker are not keeping up with productivity, it is necessary to break the gap down into component factors.

One partial explanation for the gap between productivity and median real wages is that total compensation – including non-cash benefits – has grown faster than wages, in part due to rapid increases in health insurance premiums. “Anything that just looks at cash wages is incomplete,” says Greg Mankiw, a Harvard professor and former chairman of the Bush council of economic advisers. “On the other hand, no question there is increasing inequality between high and median income people – no adjusting for compensation would change that.”

Unfortunately, there is no good and timely measure of median total compensation that includes all these non-cash benefits. At the mean average level, hourly compensation has grown steadily since 2001, but still much less than productivity.

Another partial explanation is that prices of consumer goods have risen faster than those of domestically produced goods in general, so a given increase in output productivity buys fewer additional consumer goods.

Mr Bernstein and Larry Mishel, also at EPI, estimate that these two factors account for a little less than half of the total divergence between productivity and real median wages since 2000. The other half reflects growing inequality among wage earners and a shift in the share of national income captured by labour and capital.

In the 1980s, the story of wage inequality in the US was one of the poor falling further and further behind the middle class, while the rich pulled ahead. Now the poor are keeping pace with the middle, but the rich keep pulling away.

Thomas Piketty, a professor at Paris Jourdan Sciences Economique, and Em­manuel Saez, a professor at Berkeley, estimate that the share of total income captured by the top hundredth in the US doubled from 8 per cent in 1980 to 16 per cent in 2004. In fact, the gap between the middle and upper income brackets grew more rapidly in the late 1990s under Mr Clinton than it has subsequently under Mr Bush. The difference is that in the late 1990s the rising tide in the labour market was so strong that it lifted all boats, even if some rose a lot higher than others. In the early 2000s some remained stranded while others went up.

Mr Mankiw says: “Nothing special happened since 2000. Increasing inequality happened some time in the 1970s.” The late 1990s was the “exceptional period” – with real wage gains across the income spectrum – but only because the economy was “going through a bubble”. This “was not something that could be sustained”.

Democrats, however, claim labour market gains under Mr Clinton were largely the result of sound economic policies, including the elimination of the budget deficit, which encouraged business to hire.

Labour share of national income during economic recoveries

A fundamental feature of the late 1990s economy was a dramatic shift in the share of national income going to labour, which surged from 56.7 per cent in 1997 to 58.2 per cent in 2000 – and then fell dramatically from 2001 onwards, reaching a low of 56.8 per cent in 2005. Meanwhile, profits soared, reaching 13.6 per cent of GDP in the second quarter of 2006 – close to an all-time high.

Mr Lazear argues that the US is today where it was in 1997 – when the labour/capital share began to swing back towards workers. Nominal wage growth has accelerated this year – certainly at the average level, probably at the median too – with an extra boost to real incomes recently from falling energy prices. “We have seen the turnround,” he says.

Others are not so sure, particularly since unemployment is widely expected to rise from its low of 4.6 per cent, reducing pressure on companies to compete for workers with pay rises. “I hope Lazear is right,” says David Autor, an associate professor at MIT. “I don’t think it is at all normal five years out of recession for there to be this divergence between productivity and real wage growth.”

It is possible that over the next few years labour as a whole will capture a bigger share of income from corporate profits. It is also possible that it will not, since globalisation may have permanently changed the relative bargaining power of capital and labour in the industrialised world.

But even if average wages start to catch up with productivity gains, median wages may grow at best sluggishly, reflecting an increasing gap between the average American and the high-salaried elite. Glenn Hubbard, dean of Columbia business school and another former chairman of the Bush council of economic advisers, says most explanations for this revolve around the “familiar culprits – globalisation and technology”.

Increased global competition has eaten away the economic “rents”, or excess returns, earned by US manufacturing workers, he argues. Meanwhile, the growth of global corporations and markets allows “superstars” – whether in business, finance, sports, law or entertainment – to apply their talents across a much bigger base, increasing the economic return to their skills.

US real median income

The most potent force, though, may be technology rather than globalisation – though the two are inextricably linked. Larry Katz, a Harvard professor who worked in the Clinton administration, says information technology is essentially “complementary to workers at the top, a substitute for workers in the middle” and of minimal relevance to those at the bottom of the income scale.

“The question is: what’s your remedy?” says Mr Hubbard. Most economists favour increased investment in education; Mr Hubbard thinks the government should fund personal retraining. accounts for workers.

Yet education is likely to be only half a solution, since the big increase in financial rewards to education is at the very top – to MBAs and law or medical degrees – and these are not easily spread through the population. Many pro-Democrat economists favour higher taxes at the top to redistribute the economic gains more widely.

The difficulty that orthodox economists have in coming up with effective ways to spread the benefits of growth opens the door to protectionist populism. This month, median wage stagnation is a problem for the Republicans. Longer term, though, it may prove an even bigger political challenge for the Democrats. This will be the subject of a feature in Friday’s FT.

Ever harder to grow rich

A growing gap between middle and high income earners has for three decades been a pretty consistent feature in the US, writes Chris Giles. But evidence from other advanced countries suggests this is by no means an inevitable outcome.

Research by Andrea Brandolini of the Bank of Italy and Timothy Smeeding of Syracuse University and the Luxembourg Income Study shows the US alone has had high and rising inequality consistently from the mid-1970s. While the UK matched the US rise in inequality in the 1980s, the trend stopped in the early 1990s, well before the Labour government came to power.

Nordic countries have experienced a modest rise in inequality, while in France and Italy it has been falling.

The authors say the US “is an outlier among rich nations, and only Russia and Mexico, two middle-income economies, have higher levels of inequality”. That stems not so much from the US labour market, they add, but rather that its government redistributes far less through the tax and social security system than other countries do.

Definitive statements on international comparisons are hampered by a lack of directly comparable data. Most of the internationally comparative figures relate to measures of inequality of incomes rather than the growth of median earnings. But given America’s greater than average prosperity, other countries might not have enjoyed rising median wages or incomes if they had started at the same high level.

Direct evidence on wage levels is rarer still. However, Emmanuel Saez of Berkeley along with others has used tax return data to study the shares of income earned by the rich in different countries. His findings are stark.

If 1,000 people were chosen at random in the US, the UK and France in 1963, the richest person would have commanded 2 per cent of the combined income – 20 times more than if wages were equal. At the end of the 1990s, the share of the richest person had grown to 6 per cent of total income in the US and 3 per cent in the UK but was still 2 per cent in France. This suggests strongly that other countries have by no means shared the US experience in the richest appropriating the bulk of rising prosperity.

These estimates are just snapshots of incomes. So perhaps a better barometer of a society’s earnings is whether people on lower incomes have a decent chance of becoming rich. Mark Pearson of the Organisation for Economic Co-operation and Development suggests the answer is not encouraging for the US. In studies of intergenerational income and earnings, countries such as the US with high income inequality tend to have less social mobility than countries with greater equality. The son of a rich US man is much more likely to be rich than the son of a rich Swede. The same is true for a poor American versus a poor Swede.

In a Warwick University paper, Bernt Bratsberg and others show important nuances in this view of intergenerational earnings mobility. While they still find the “American dream” to be a more common reality in Nordic countries than in the US, they show big differences for different types of families in different countries.

High-earning parents in the US, the UK and Nordic countries are all fairly good at ensuring their children also become rich. Rich US parents are particularly skilled in this feat. In the US and to a lesser degree the UK, sons of poor fathers are disproportionately likely to be poor. But in Nordic countries this is not true. The sons of the poor are no less likely to succeed than those of middle-income fathers.

So while the poor have a very realistic expectation of becoming middle class in Scandinavia, it is an exception to the rule in the US. Median wage stagnation in the US threatens to expose this truth.

If US citizens think this lack of mobility is a fair and meritocratic outcome, reflecting the value of different families, the lack of social mobility is no great problem. But to the extent that Americans believe rigidities in their society impede success, unhappiness is likely to mount.

Get alerts on US downturn when a new story is published

Copyright The Financial Times Limited 2019. All rights reserved.
Reuse this content (opens in new window)

Comments have not been enabled for this article.

Follow the topics in this article