New Economist

Why people emigrate

Semi-regular blogging service resumes this week with a few posts on migration - still a very topical issue on both sides of the Atlantic.

The first paper I'd like to highlight is by the University of Chicago's Jeffrey Grogger, and UCSD's Gordon H. Hanson. Their recent NBER Working Paper No. 13821, Income Maximization and the Selection and Sorting of International Migrants, seeks to explain to what extent selection and sorting account for international migration flows using data on emigrant stocks by schooling level and source country in OECD destinations. As the authors conclude, a simple model can explain a lot:

Two dominant features of international labor movements are positive selection of individuals into migration and positive sorting of migrants across destinations. We show that a simple model of income maximization can account for both phenomena.

The more educated are more likely to emigrate; and more-educated migrants are more likely to settle in destination countries with higher rewards to skill. As the authors explain:

In our selection regression, we find that migrants for a source-destination pair are more educated relative to non-migrants, the larger is the skill-related difference in earnings between the destination country and the source. That is, positive selectivity is stronger where the reward to skill in the destination is relatively large. This result obtains for wage differences expressed in levels, but not in logs.

...Positive sorting is a general prediction of income maximization. In our sorting regression, the relative stock of more-educated migrants in a destination is increasing in the level earnings difference between high and low-skilled workers. This correlation is stronger when wage differences are adjusted for taxes, implying that migrants weigh post-tax earnings when choosing a destination. The U.S. and Canada enjoy relatively large post-tax skill-related wage differences, which largely account for their ability to attract more educated migrants relative to other OECD countries.

Other factors are also at work:

Our analysis also shows that language, history, and policy affect migration. English-speaking destinations draw higher-skilled immigrants than other destinations, whereas former colonial powers draw lower-skilled immigrants from their former colonies than from other source countries. Destinations with liberal refugee and asylum policies draw relatively low-skilled immigrants, all else equal.

An ungated version of the paper is available here or here.

Globalisation: good for jobs?

Chris Giles summarises a new report on globalisation by the Ifo-affiliated European Economic Advisory Group in today's Financial Times: Globalisation ‘a blessing’ for west Europe

Increased trade, outsourcing and offshoring do not create unemployment but boost the number of jobs in advanced economies, a study of European labour markets says on Tuesday. The European Economic Advisory Group...argues that although globalisation can lead to a fall in demand for certain types of skill, it also tends to sweep away job-destroying rigidities in labour markets.

The evidence from the group’s work suggests the positive effects of globalisation outweigh the negative effects. Although the group concedes that its statistical work remains “crude”, the report concludes that globalisation is likely, if anything, to lead to long-term rises in employment. “If so, globalisation will not be a curse for employment in western Europe, it could instead turn out to be a blessing,” the report says.

The report in question is Chapter 3 of the EEAG's seventh report on the European Economy, Europe in a Globalised World, launched today in Brussels. It warns of "dark clouds" in the US, but no recession in Europe. The chapter, The effect of globalisation on Western European jobs: curse or blessing? (PDF 2.9Mb), comes to more qualified and tentative conclusions than the FT article suggests:

Our basic message is that we probably should not expect globalisation to have adverse effects on overall employment in Western Europe in the long run if one takes all effects into account. It is true that trade integration and factor mobility vis-à-vis low-wage economies are likely to cause unemployment if European labour markets remain rigid. But there is a good chance that globalisation will help reduce these rigidities. Politicians in some countries may try to swim against the tide and uphold or even strengthen regulations in the labour market, such as Germany is currently doing. But in the end, globalisation is likely to strengthen the incentives to deregulate. Therefore, the net result could be that employment is promoted.

If globalisation does not hurt employment, it will produce aggregate gains. There is a possibility that globalisation could eventually benefit almost everyone, although some will gain more than others. However, there is also a fair amount of evidence that economic integration with low-wage economies reduces the relative demand for less-skilled workers and their relative compensation. So, it is also possible that there could be a large group of losers.

But maybe that's the wrong way of looking at it? The authors suggest we should examine how effectively our institutions handle a more global world:

It makes more sense to recast the issue in the following way: are our labour market institutions and our welfare states designed well enough so that the gains from trade reform will be broadly shared? Or are they likely to breed opposition to these reforms?

The 34 page chapter ends with a useful discussion of the possible components of schemes to compensate the potential losers from the globalisation process. A thoughtful piece, deserving a wide readership.

The mixed benefits of remittances

Recent years have seen international agencies like the World Bank and IMF extol the econmic benefits of remittances. Sending money to the folks back home boosts the incomes of developing countries and helps to offset losses fom the 'brain drain'. What's not to like?

Well, we wouldn't be economists if we weren't looking for unintended consequences. And sure enough, David Grigorian and Tigran Melkonyan have found some. Their recent IMF working paper Microeconomic Implications of Remittances in an Overlapping Generations Model with Altruism and Self-Interest studies the impact of remittances on Armenian households:

We demonstrate that when the migrant and the relative(s) cooperate to maximize the joint utility of the household, this leads to higher level of remittances as well as investment and hours worked by the relative(s). We use data from Armenia to test our predictions regarding implications of remittances flows on behavior of receiving households.

Consistent with our predictions, remittance-receiving households work fewer hours and spend less on the education of their children. While saving more, these households are not leveraging their savings to borrow from the banking system to expand their business activities. This evidence suggests that the benefits of remittances might be overstated and emphasizes the importance of measuring their impact in a general- rather than a partial-equilibrium context.

So remittances lead to households working fewer hours: "The coefficient is negative and significant and its magnitude is rather large." More surprising is the apparent negative effect on education. The authors conjecture:

The impact of remittances on education spending (column 3, Table 3) is perhaps the most controversial of our findings. The negative (and significant) coefficient here could be indicative of two things. First, it is possible that members of remittance-receiving households are likely to later migrate themselves and, therefore, not value the local education as much. Second, because their consumption patterns might be under scrutiny by the remitter, the receiving households may adjust their consumption pattern to look more conservative and be
centered around necessities (such as food and public services/utilities, and presumably not education and other types of spending that could be considered unnecessary from theremitter’s point of view). To the extent that remittances represent a large share of the receiving family’s income, for the same level of disposable income, this tendency to “simplify” the spending pattern could in fact lead to lower spending on education (in nominal terms) out of total income.

Families with remittances do accumulate more savings. While the costs of migration on households have a negative effect, it is "not large enough ..to offset the accumulation of savings due to remittances."

I would be interested to see if similar remittance effects are to found in other countries with high levels of emigration.

Britain's super rich: racing away?

The latest Institute for Fiscal Studies briefing note, Racing away? Income inequality and the evolution of high incomes, focuses on the 'super-rich'. Authors Mike Brewer, Luke Sibieta and Liam Wren-Lewis define this group of 'high income individuals' as the richest 1 in every 1,000 taxpayers (i.e. the top 0.1%) The super-rich earn £350,000 or more in taxable income a year. Nine in every ten are men, and most are lawyers, City bankers and the like. The accompanying IFS press release states:

The outlook for inequality in Britain may depend more on the outlook for the stock market than on Government tax and benefit policies, a study by IFS researchers suggests today. Even though the current Government has increased taxes on people with high incomes, this has not prevented them from them racing further away from the average level of living standards across the country. In recent years, it is only in the wake of extended falls in the stock market that the incomes of the richest have fallen.

The FT's Chris Giles comments on the role played by market forces in his piece, Very rich get richer under Labour:

Highlighting a sea-change from the early post-war years when wealth often derived from land or other inherited assets, Mr Sibieta said the 21st century super-rich received 80 per cent of their incomes from an occupation - whether salaried or self-employment - rather than investments. "We are talking about the working rich rather than the idle rich," said Mr Sibieta.

But the past decade has seen rich people's income sputter as well as soar. With so many working in finance, there is a strong link between their fortunes and those of the stock market. Real incomes of top-earners grew 6.6 per cent a year on average between 1996-97 and 2001-02.

Yet in the following two years, when the stock market reached its low point, they fell on average by 2.7 per cent, before picking up again alongside equities in 2004-05. That tight correlation leads the IFS to predict that, with the rapid growth of financial markets since 2004-05, the incomes of the richest will have risen quickly in the past two years.

The data show the impact of Labour's generous funding of the public services this decade. The one group outside finance, the law or property to be well represented in the top 1 per cent of incomes works in the health sector. These were "presumably doctors and senior health service managers who have enjoyed relatively big pay increases under Labour", added Mr Sibieta.

The 'super-rich' earn 31 times average incomes: around £780,000 pre-tax income a year, of which they pay on average almost £275,000 in income tax: a tax rate of 35.3%. This is double the rate paid by the average UK taxpayer, who earns around £25,000 and pays £4,400 in tax: a rate of 17.6%.

How useful is Okun's Law?

Edward S. Knotek asks an interesting question in the latest Kansas City Fed's Economic Review: How Useful is Okun's Law? (PDF)

From the beginning of 2003 through the first quarter of 2006, real gross domestic product in the United States grew at an average annual rate of 3.4 percent. As expected, unemployment during the period fell. Over the course of the next year, average growth slowed to less than half its earlier rate--but unemployment continued to drift downward. This situation presented a puzzle for policymakers and economists, who expected the unemployment rate to increase as the economy slowed.

Typically, growth slowdowns coincide with rising unemployment. This negative correlation between GDP growth and unemployment has been named “Okun’s law.” Part of the enduring appeal of Okun’s law is its simplicity, since it involves two important macroeconomic variables. Additionally, the relationship appears to enjoy empirical support. In reality, though, Okun’s law is a statistical relationship rather than a structural feature of the economy. As with any statistical relationship, it may be subject to revisions in an ever-changing macro economy.

Knotek considers the usefulness of Okun’s law for policymakers and economists. The evidence suggests that Okun’s relationship between changes in the unemployment rate and output growth has varied considerably over time and over the business cycle. Nevertheless, Okun’s relationship can still be useful as a forecasting tool--provided that one takes its instability into account.

Hollywood and the cost of revealed talent

Why are there more mediocre films being made by Hollywood now than during the 'golden years' of the studio system in the 1930s and 1940s? Drawing on a new paper by Berkeley's Marko Tervio, Chris Dillow at Stumbling and Mumbling offers a plausible explanation in his post Superstars and talent:

Top bosses and film stars get multi-million pound salaries because talent is scarce. Everyone knows this. Which is a shame, because it's bull, as this fantastic paper by explains.

Start from the premise that talent is initially unknown, and can only be revealed by working with expensive equipment. So, for example, we can only find out if a manager is any good if he's in charge of a big venture, or if an actor has box office appeal if he's in a mega-costly film. It is, therefore, very expensive to learn who's got talent and who hasn't.

What's more, people with talent cannot offer to share this cost with employers, either because of lack of cash or risk aversion: people don't pay for the chance to become bosses or film stars.

In these conditions, what's scarce isn't talent, but revealed talent. There might be loads of people with the ability to be film stars or bosses, but only a handful get the chance to show what they can do. Marg Helgenberger gets big money not (just) because she's a better, more popular or more beautiful actress than others, but because she's a proven quantity.

This has three consequences:

1. The industry employs lots of mediocrities on good money - people just above the threshold of acceptable competence. Employers prefer these to untried but potentially better workers because the risk of a single failure - the box office flop of a $100m film, or the failure of a large company - far exceeds the possible benefit of finding lots of people a little better than the barely competent.
2. Output is inefficiently low, and prices high. This follows from the industry being staffed by the barely competent rather than the brilliant.
3. People who have proved that they are genuinely brilliant earn huge salaries as economic rent.
Tervio illustrates this with Hollywood films. Compare films today with those made by the studio system of the 1930s and 40s.

Today, there's vast inequality among film actors, as there's only a handful of proven box-office stars, and the quality of films is indifferent.

By contrast, the studio system allowed potential stars to share the costs of revealing their talent with producers, by entering long-term contracts which gave low pay even to stars; the lower cost of making films then also helped. This allowed for many more great films to be made, as more great actors could enter the industry, but with less inequality among the stars.

In this sense, superstar salaries arise not because talent has become scarcer, but because it's become harder to reveal talent.

In other industries, however, the opposite trend exists. Clive reminds us that in the 1980s a mere economics writer got a flash Beemer from the Indy. This would be unheard-of today. This is because back in the 80s, it was hard to reveal a "talent" for writing on economics. Today, any idiot can do this (see here, passim), so revealed talent has become more abundant.

Marko's paper, Superstars and Mediocrities: Market Failure in the Discovery of Talent (PDF), is well worth reading.

Obese American women's rising wage penalty

The latest foray into the economics of obesity debate comes from David Lempert of the Bureau of Labor Statistics. His new economic working paper, Women's Increasing Wage Penalties from Being Overweight and Obese, finds that overweight working women in the United States face a rising wage penalty:

This paper first utilizes annual surveys between the 1981 and 2000 waves of the National Longitudinal Survey of Youth to estimate the effect of being overweight on hourly wages. Previous studies have shown that white women are the only race-gender group for which weight has a statistically significant effect on wages. This paper finds a statistically significant continual increase in the wage penalty for overweight and obese white women followed throughout two decades.

A supporting analysis from a cross-sectional dataset, comprised of the 1987 National Medical Expenditure Survey and the 2000 and 2004 waves of the Medical Expenditure Panel Survey, also shows an increasing wage penalty. The bias against weight has increased, despite drastic increases in the rate of obesity in the United States. Alternatively, the increasing rarity of thinness has led to its rising premium.

Lempert concludes:

The increasing wage penalty corresponds to current psychological research that demonstrates increased weight stigmatization in the United States. Further, as larger women age, their wages incur the effects of years of cumulative discrimination. With other factors controlled, their starting wages are lower. Throughout their working careers, these women receive less frequent raises and promotions. Therefore, we see increasing penalties in both NLSY data and the synthetic cohort constructed from NMES/MEPS data.

This paper has shown that an obese 43 year-old woman received a larger wage penalty in 2004 than she received at 20 in 1981. This paper also provides some evidence that an obese 20 year-old woman receives a larger wage penalty today than she would have in 1981 at age 20. Future literature should further explore this aspect of the story, as well as the mechanisms by which the wage disparities occur. It can be concluded that increased body weight has drastic economic consequences that have grown over time.

This is a surprising finding. One would have thought it was now the norm to be overweight in the US, and employers have become habituated to it. This paper provides strong evidence to the contrary. Comments welcome.

Is there a 'marriage premium' for gay men?

No, according to a new IZA discussion paper by Madeline Zavodny:

Controlling for observable characteristics, cohabiting gay men do not earn significantly more than other gay men or more than unmarried heterosexual men.

But she also finds that:

Cohabiting heterosexual men also do not earn more than non-cohabiting heterosexual men.

Using US data Zavodny estimates that the marriage premium, which accrues only among heterosexuals, is about 18%; this is in line with previous estimates in cross-sectional data.

Why does marriage matter? One possible answer, proposed by Becker, is that it encourages specialisation within couples. But the authors findings seem inconsistent with that hypothesis:

The results indicate a positive relationship between a man’s earnings and the education of his spouse/partner. This is similar to previous research and consistent with both a positive productivity effect and positive assortative mating.

...spouses/partners reduce their own labor supply if their husband/partner earns more. Controlling for this effect by using predicted hours yields results that are more consistent with positive assortative mating than with specialization for all three types of couples.

These results appear consistent with a recent paper by Alison Booth and Jeff Frank of academics and administrators at British universities: Marriage, Partnership, Cohabitation and Sexual Orientation: What Males Gain a Wage Premium? (PDF):

We find a statistically significant male marriage premium, an insignificant positive effect of heterosexual unmarried partnership, and no partnership return to male homosexuals. This suggests that selection may play a limited role in the marriage premium. We also provide results on cohabiting versus non-cohabiting partners, and on the academic versus administrative side of universities.

UPDATE:
Tyler Cowen from Marginal Revolution comments:

Data on cohabitation suggest that the answer is no, whether for gay men or cohabiting heterosexuals.  The standard selection story is that women are more likely to choose the high earning men and marry them.  But why don't women live with these men too?  Does living together not transfer enough resources?  Could it be that real legal marriage is proxying for the ability to commit, which is positively correlated which other determinants of job success?

See his post for comments.

In conversation with Richard Freeman

One of America's most prolific and well respected economists, Harvard's Richard Freeman, talks about his career and his research at UC Berkeley, available on YouTube: Global Capitalism, Labor Markets, and Inequality

Hat tip: BookForum.com

More from the UK migration debate

With not just tabloids like the Daily Mail but even government Ministers emphasising the potential downsides of the UK's recent migration experience, a few commentators have sought to present more thoughtful accounts. Here are three:

David Smith of the Sunday Times wrote on 28 October that Migrants ease the inflation pressure. he does a good job summarising submissions to the House of Lords economic affairs committee investigation into the economic impact of immigration:

There are contributions from business, with the Institute of Directors saying that three-quarters of its members believe that foreign-born workers make a big positive contribution to the economy and, more controversially, that they “significantly outperform the existing workforce across a whole range of measures, including productivity, education and skills, work ethic, reliability and the amount of sick leave”.

The City of London Corporation waxes lyrical about the historic contribution of immigrants, noting that “City street names such as Lombard Street date back to the reign of King Edward I, when land was given to goldsmiths from the Lombardy region in Italy”. More recently there has been a significant City influx from the Continent and America and the corporation notes that London’s overall foreign-born population rose from 1.17m to 2.23m between 1986 and 2006.

J Sainsbury says it has used migrant labour to fill “pressing gaps in both skilled and unskilled areas” and praises the work ethic of such staff.

This kind of thing would be a red rag to the unions, it might be thought. But the Trades Union Congress, in its submission, is also strongly supportive.

“The overall economic impact of immigration is limited but positive,” it says. “Migrant workers contribute more in taxes than they receive in services, and migration probably leads to slightly higher levels of employment and wages for native workers. Migration may possibly be linked to an increase in wage inequality in this country, but the evidence is not conclusive.”

The government’s take on the economic impact of immigration is also reassuring. It believes that in the short term the impact of immigration on the public finances is positive, though the last official study was some years ago. It concluded that migrants paid £2.5 billion more in taxes than they took out in benefits and the use of public services. This net gain to the exchequer is likely to have grown, the government suggests.

The long-term impact of migrants is more difficult to assess, it accepts. An American study pointed to a large overall positive impact but any conclusion for Britain is sensitive to the number of dependants each migrant has, and whether they remain in Britain until retirement. As it is, the effect of migration will be to ameliorate the consequences of Britain’s ageing population.

The dependency ratio – the ratio of dependants to working-age people – is 61% and will rise to 74% over the next 50 years. With zero net immigration it would increase to 82%.

The arrival of migrant workers has meant employers have not had to outbid each other for scarce labour. This has helped the Bank of England by delivering lower wage rises for a given rate of economic growth than would otherwise have been the case. Interest rates have thus been lower in recent years than they might have been.

The IPPR's Danny Sriskandarajah explains in the Financial Times why limiting immigration is bad for Britain. He points out how difficult it would be to 'control' migration, and argues that work permit system is likely to prove more responsive than a points-based system.

In their rush to control immigration, both parties are also suggesting a departure from the market-based, employer-led system of allocating work permits that has served the UK economy so well over recent decades. ..the number of work permits issued each year has broadly been in keeping with economic and labour market conditions. An individual-led, points-based system, in which someone need not have a job offer in the UK, may not be sufficiently responsive.

Annual limits or sector-based quotas are likely to be even worse. Imagine a desperate would-be employer being told that the new head of mergers and acquisitions or eminent university professor or star footballer cannot take up his or her post until next year because the annual limit has already been exceeded.

Britain’s political leaders either do not realise that migration cannot be cut in the crude ways they say they want to, or they have calculated that political benefit of being tough on immigration from outside Europe is worth the undoubted economic costs this may incur.

Unfortunately, denying UK employers access to the best and brightest workers from across the world may hurt more than they expect and still not stop the inflows that worry the electorate.Indeed, without a long-overdue reality check, Britain may continue to have relatively high levels of immigration but not necessarily the right workers for the right jobs.

I am less hostile to the new points system than Danny, but agree that attempts to curb migration may well backfire.

Finally, Richard Reeves examines the impact of immigration in this week's New Statesman: Why Brits need not apply