Photo by Bayat on Flickr

Two odd pieces of economic research have been highlighted in the Ig Nobel Prizes, awarded each year by the magazine the Annals of Improbable Research.

The first study examined tips given to lap dancers. Unfamiliar with lap-dancing, Geoffrey Miller and colleagues read up on the relevant sociological and feminist literature before getting eighteen dancers to record their earnings for two months. They found that earnings were greater when the dancers were ovulating: the male patrons expressed a preference for dancers who were currently fertile, even if not consciously aware of the difference.

The other study was by behavioural economist Dan Ariely and colleagues, who found that the placebo effect of a pill was weakened when the pills were discounted in price. In other words, some medicines are more powerful in virtue of being more expensive.

The full references are Geoffrey Miller, Joshua M. Tybur, Brent D. Jordan (2007) “Ovulatory Cycle Effects on Tip Earnings by Lap Dancers: Economic Evidence for Human Estrus?” Evolution and Human Behavior, vol. 28, 2007, pp. 375-81; and Rebecca L. Waber, Baba Shiv, Ziv Carmon, Dan Ariely (2008) “Commercial Features of Placebo and Therapeutic Efficacy” Journal of the American Medical Association, March 5, 2008; 299: 1016-1017.

Listen to the interview

Download audio file (tominey128.mp3)

Mothers who smoke during pregnancy will have smaller babies. But much of the harm is due to unobservable traits of the mother. If mums stub it out by the time they are five months pregnant, the damage is as good as undone.

At the same time, the lasting harm to babies is greatest if the mothers have low education. So a much more holistic approach to improving child health in pregnancy is needed to help thousands of children break out of the poverty trap.

These are the conclusions of extensive new research by Emma Tominey, presented to the Royal Economic Society’s 2007 annual conference at the University of Warwick, 11-13 April.

Babies born to women who smoke will typically be 5.4% (6.5oz) lighter than other babies. But around half of this damage is because of unobservable traits of the mother. This means that stopping mothers smoking during pregnancy is important, but it is only half of the battle.

So while the effects of being a small baby stay with a child throughout its life, affecting its health, education and earnings potential, stopping a mother from smoking must be combined with helping her to be healthier in other areas of her life.

But for the harm that remains, the low educated mothers are hardest hit. Children born to mothers who left school at the age of 16 suffer double the harm for each cigarette smoked. The government must target its policy directly at these low educated families.

Women who do smoke in the early stages of pregnancy should not be ‘written off’ as ‘too late’. Surprisingly, the research shows that the harm to the baby is essentially reduced to zero if the mother quits by month five of the pregnancy.

This is much longer than conventional wisdom and previous research has suggested and tells us there’s more time than we thought to help the mothers change their behaviour during pregnancy.

The study is based on research into the lives of 6,500 children and their mothers, and went into exceptional detail of the mother’s lifestyle over her lifetime. The mothers were tracked from their child’s birth until the age of 42.

The research suggests that while previous studies have identified a link between smoking and low birth weight, none has looked in such depth at whether the experiences of the mother can alter this and how the harm accumulates during pregnancy.

The study calls on the government to alter radically its policy on helping pregnant women quit smoking, developing a more holistic approach to improving the health of these children during pregnancy and targeting the children of low educated mothers.

Notes for editors: Maternal Smoking During Pregnancy and Child Birth Weight by Emma Tominey was presented at the Royal Economic Society’s 2007 annual conference at the University of Warwick, 11-13 April.

Emma Tominey is at University College London.

For further information: contact Romesh Vaitilingam on 07768-661095 (email: romesh@compuserve.com).

Read more papers by Emma Tominey at EconPapers and search for more Internet resources on the issue of Women and Economics at Intute: Social Sciences.

Listen to the interview

Download audio file (prowse128.mp3)

No matter what qualifications they have or how big their family is, British women face a substantial occupational penalty if they work part-time. That is the central finding of new research by Victoria Prowse, presented to the Royal Economic Society’s 2007 annual conference at the University of Warwick.

This finding is consistent with an inadequate supply of high quality part-time jobs to suitably qualified women, and provides support for the introduction of incentives for the firms to increase the number of part-time jobs in skilled occupations.

The study also finds that women with children who enter the labour market have higher occupational attainment, and experience a smaller occupational penalty, than childless women. Thus, there is no evidence whatsoever of women with children (the majority of whom work part-time) being any less career-focused than women without children.

It is well known that part-time jobs in the UK are concentrated in poorly paid, low skilled occupations such as catering and retail. This study estimates that between 1974 and 2000, an average of 76% of women in full-time jobs were working in non-manual occupations while on average only 56% of women in part-time jobs were working in non-manual occupations.

The study draws on data from the National Child Development Survey in order to build up a picture of the employment choices and occupational attainment of British women between 1974 and 2000. It finds that:

• Holding a university degree reduces, but does not eliminate, the part-time occupational penalty.
• Women with children who enter the labour market have higher occupational attainment, and experience a smaller occupational penalty, than childless women.
• But irrespective of qualifications or family size, all women experience a significant part-time occupational penalty.

A university degree reduces the occupational penalty suffered by women in part-time work: among women aged 24 with no qualifications, those in full-time work are 27 percentage points more likely than part-timers to be working in a non-manual occupation.

In contrast, among women holding university degrees, those in full-time work are only 18 percentage points more likely to hold a job in a non-manual occupation. Therefore, although highly qualified part-timers suffer a substantial part-time occupational penalty, it is far less than that suffered by women with lower levels of qualifications.

The presence of children in a woman’s household also has implications for her occupational attainment. Women with children are more selective in terms of the quality of jobs that they are willing to accept than childless women: conditional on being in either full- or part-time employment, women with children are on average 10 percentage points more likely than childless women to be employed in non-manual occupations.

Thus, there is no evidence whatsoever of women with children being any less career-focused than women without children. Interestingly, this selectivity effect is greater for women in part-time jobs than those in full-time jobs. So the low occupational attainment of women in part-time work cannot be attributed to low occupational ambitions among women with children, the majority of whom work part-time.

Despite variation in the part-time occupational penalty across women with different levels of education and with different family sizes, all women experience a substantial part-time occupation penalty: after controlling for differences in individual characteristics, women in part-time work are on average 14 percentage points less likely than full-time workers to employed in non-manual occupations.

This finding is consistent both with the presence of a constraint on the supply of high quality part-time jobs and with women in part-time employment having a strong preference for jobs in low occupations. To the extent that there is a constraint on the supply of high quality part-time jobs to suitably qualified women, there are grounds for policy interventions aiming to equalise the occupational opportunities of women in full- and part-time employment.

Notes for editors: Part-time Work and Occupational Attainment Among a Cohort of British Women by Victoria Prowse was presented at the Royal Economic Society’s 2007 annual conference at the University of Warwick, 11-13 April.

Victoria Prowse is at the University of Oxford.

For further information: contact Romesh Vaitilingam on 07768-661095 (email: romesh@compuserve.com).

Find more papers by Victoria Prowse at EconPapers and search for more Internet resources on the issues of women and employment and the labour force and market at Intute: Economics.

Listen to the interview

Download audio file (walker128.mp3)

Academics are underpaid and overworked compared with other graduate professions and this is likely to have consequences for the quality of UK degrees. That is the conclusion of new research by James Walker and colleagues, presented to the Royal Economic Society’s 2007 annual conference at the University of Warwick.

The study finds that:

• Not only do academics work longer hours than the average graduate, but they also earn around 3% less.
• In terms of their earnings, academics compare particularly poorly with accountants, those in the legal professions, consultants, engineers, physicians, pharmacists and dental practitioners (across both the public and private sectors).
• On average, academics earn approximately 17% less than other similarly qualified individuals in the accountancy profession, 23% less than lawyers, 24% less than doctors and 49% less than dentists.
• Only two groups of workers do worse than academics: teachers in further education and, to a lesser extent, secondary school teachers.
• Academic pay is an important policy issue because if the relative pay of academics falls, it is likely to lead to lower quality individuals entering and remaining in the profession, as well as a brain drain to countries that reward academics more highly.
• These trends are in turn likely to have a knock-on effect on the quality of UK higher education.

Higher education is of crucial importance to the UK economy. The sector is estimated to be worth £45 billion. It is also becoming a global business with current export earnings of about £3.6 billion. The sector is playing a key role in achieving the government aspiration for a high skill strategy to transform the UK into a knowledge economy, as reflected in the target of 50% participation in higher education by the end of the decade. Maintaining the quality of the sector is vital for preserving its high international status and for producing high quality skilled labour for the economy.

Economic theory would suggest that if academic pay falls, relative to other similar professions, the calibre of individuals entering and remaining in the profession is likely to fall. But robust analysis of relative academic pay in the UK has been limited.

James Walker and his colleagues have analysed the size of the gap in hourly earnings between academics and graduates in a range of other comparable occupations. The study identifies the gap in hourly earnings between different occupations after taking account of a range of other differences in the characteristics of individuals who go into different professions, such as their gender, ethnicity and of course their education level.

The study concludes that after taking account of these other factors, academics still earn somewhat lower earnings than most public sector and private sector graduates and that they do particularly poorly compared with a range of comparable professions.

Of the other graduate professions considered by the authors, dentists are the best paid followed by doctors, lawyers, accountants, legal professionals, accountants, pharmacists and pharmacologists, consultants and engineers; all of whom earn considerably more then the average graduate. Indeed, of the ten groups examined, it is only the three education professions  academics, teachers in further education and secondary school teachers who are paid less then the average UK graduate.

Not only are academics relatively lowly paid compared with other professions but they also work longer hours then other graduates. While the average graduate worked around 44 hours per week in 2004, academics worked 47 hours per week. Indeed, with the exception of doctors who worked around 51 hours per week in 2004, academics work longer hours than the other ten professions.

The study was not able to consider the non-pecuniary benefits from working as an academic, such as the ability to do flexible working hours or have greater intrinsic interest in the work. Further research is needed to determine whether these non-pecuniary benefits are likely to compensate for lower wages.

The results from the study suggest that policy-makers should be concerned about the relative low pay of academics. As pay in the higher education sector is largely not determined by a free labour market, there is a risk that relative wages for academics may decline further and thereby reduce the quality of the workforce in higher education.

Notes for editors: Higher Education Academic Salaries in the UK by James Walker, Anna Vignoles and Mark Collins was presented at the Royal Economic Society’s 2007 annual conference at the University of Warwick, 11-13 April.

James Walker is at the University of Reading Business School; Anna Vignoles is at the Institute of Education.

For further information: contact Romesh Vaitilingam on 07768-661095 (email: romesh@compuserve.com).

Find more Internet resources on the issue of the economics of education at Intute: Social Sciences.

Listen to the interview

Download audio file (papps128.mp3)

Married women work more hours in the labour market when they face a high likelihood of divorce: for example, a woman who is unhappy with her marriage will work on average 283 hours more in the following year than a woman who is very happy with her marriage. In contrast, married men are unaffected by the probability of divorce.

These are among the findings of new research by Kerry Papps. The study also finds that both single men and single women work more when they have a high chance of marrying in the near future.

These findings are generally consistent with the idea that people take account of their future wellbeing and specialise when they are married by working either in the labour market or at home. If a married woman who maintains a household believes that divorce is impending, she will wish to enter the labour market or increase hours of paid work, because this will allow her to acquire more work experience and boost her potential income in the future, when she may be forced to rely on her own resources.

Labour force participation among married American women increased steadily for four decades from 1950, largely due to major increases in the wage rate paid to women. During the 1990s, however, this growth stalled, despite unprecedented wage gains for married women over the decade.

At least part of this and other inconsistencies in the wage explanation is likely explained by changes in divorce rates, which peaked in the early 1980s in the United States, along with the UK and other western countries, after two decades of increases.

This study analyses data from the National Longitudinal Survey of Youth, which focuses on a representative sample of Americans who were teenagers in 1979 and has asked a wide range of questions every one or two years since. A fundamental problem is how to measure a persons chance of marrying or divorcing, as this obviously cannot be directly observed.

The paper takes three different approaches to estimating probabilities of changing marital status, all of which yield the same conclusions:

• First, marriage and divorce rates among people with similar demographic characteristics are used.
• Second, whether or not a survey respondent changes marital status in the year following any interview is used to form probabilities of marriage and divorce. A complication with this approach is that work decisions are likely to influence the likelihood of marriage or divorce in the future, necessitating the use of statistical techniques to deal with the possibility of causal relationships in two directions. Doubling a womans probability of divorce will result in her working 60 additional hours a year, all else being equal.
• Finally, an innovation of this study is to measure of the risk of divorce by using peoples evaluations of how happy they are with their marriages. This avoids many of the problems of the other approaches. The results are statistically significant and are striking: a woman who is unhappy with her marriage will work on average 283 hours more (or 6 hours per working week) in the following year than a woman who is very happy with her marriage.

The evidence indicates that there are two distinct dimensions to the relationship between work hours and the risk of divorce. Woman in demographic groups that suffer from high divorce rates tend to work long hours over their entire lives. Meanwhile, individual married women also respond to unexpected increases in the likelihood of divorce from year-to-year by increasing the hours they spend in paid work.

These findings point to an unexpected negative consequence of a falling divorce rate and imply that this may hamper government efforts to increase labour force participation among married women. If current trends persist, ever-larger wage gains may be needed to induce more married women to enter the workforce.

Notes for editors: The Effects of Divorce Risk on the Labour Supply of Married Couples by Kerry Papps was presented at the Royal Economic Societys 2007 annual conference at the University of Warwick, 11-13 April.

Kerry Papps is at Cornell University.

For further information: Romesh Vaitilingam on 07768-661095 (email: romesh@compuserve.com).

Find more papers by Kerry Papps at EconPapers and search for more Internet resources on the issues of women and economics and labour economics.

Listen to the interview

Download audio file (milas128.mp3)

Monetary policy-makers do not make minor adjustments to interest rates when inflation is close to the inflation target  but they do respond vigorously when inflation begins to move further from the target. That is the central argument of new research by Professors Christopher Martin and Costas Milas, presented to the Royal Economic Society’s 2007 annual conference at the University of Warwick.

UK interest rates were increased in January 2007 in response to an unexpectedly high jump in inflation, to 3%. The immediate danger of higher interest rates may have passed as inflation fell below 3% in February. But the Monetary Policy Committee is known to be concerned that oil and gas prices will lead to increased wage demands; and the Governor of the Bank of England has signalled willingness to act to avoid this.

This has led many commentators to expect higher interest rates this spring. The analysis of Martin and Milas suggests that this response could be stronger than many think: further, even small, rises in inflation are likely to trigger higher interest rates.

Interest rates are often constant for long periods of time. The Bank of England base rate did not change in 2002 or between the summer of 2004 and autumn 2005. This is puzzling. The Monetary Policy Committee of the Bank of England adjusts interest rates in order to keep inflation close to the target. In these periods, inflation changed, so why where interest rates kept constant?

The argument of this research builds on the opportunistic approach to monetary policy, an analysis of the behaviour of the Federal Reserve by the Fed’s own economists. The opportunistic approach suggests that interest rates remain constant so long as inflation remains within a zone of discretion around the inflation target but are adjusted when inflation moves outside the zone (the United States does not have an official inflation target, but most observers feel the Fed aims at an inflation rate of around 2%).

This approach implies that more traditional models of monetary policy, which assume that interest rates respond proportionately to all movements in the inflation rate are wrong; they over-predict interest rate changes when inflation is close to the target and under-estimate changes when inflation is further away.

Martin and Milas estimate various statistical models of interest rates and conclude that policy-makers in the United States do indeed follow the opportunistic approach and that the traditional model is wrong. They argue that the width of the zone of discretion is 2%, so inflation is allowed to fluctuate within 1% either side of the inflation target without triggering an interest rate response. Once the zone is breached, however, the response of interest rates is stronger than previously thought.

What are the implications for the UK? The UK economy is similar to that of the United States and the UK has a similar policy-making structure. Therefore the opportunistic approach is likely to be equally applicable to the UK (other statistical analysis by Martin and Milas confirms this).

This is the likely explanation of the long periods of static interest rates in recent years. This can also explain the large rapid changes in interest rate that have occurred in recent years, for example the fall of nearly 2.5 percentage points in 9 months in the late 1990s and the fall of 2 percentage points in 2001.

But there is one important difference between the UK and the United States, which is that UK policy-makers must write an open letter of explanation if inflation deviates from the target by more than 1%. The desire to avoid this suggests that the zone of discretion in the UK is narrower, less than the 2% zone estimated for the United States.

Notes for editors: Testing the Opportunistic Approach to Monetary Policy by Christopher Martin and Costas Milas was presented at the Royal Economic Society’s 2007 annual conference at the University of Warwick, 11-13 April.

Christopher Martin is at Brunel University. Costas Milas is at Keele University.

For further information: contact Romesh Vaitilingam on 07768-661095 (email: romesh@compuserve.com).

Find more papers by Christopher Martin and Costas Milas at EconPapers and search for more Internet resources on the topic of interest rates at Intute: Economics.

Listen to the interview

Download audio file (drinkwater128.mp3)

The typical Asian working age male is now younger, better educated and more likely to be UK-born than his parents generation were. According to research by Ken Clark and Stephen Drinkwater, each of these factors contributes to lower rates of self-employment, particularly among men of Indian and Chinese ethnicity. This suggests greater assimilation of these groups into the UK labour market and education system.

But the study, presented to the Royal Economic Society’s 2007 annual conference at the University of Warwick, also finds relatively stable rates of self-employment among Pakistani and Bangladeshi men. It seems likely that discrimination in paid employment against these groups is keeping them in self-employment, working long hours in relatively poorly rewarded sectors such as catering and taxi-driving

The stereotype of the Asian businessman, often a shopkeeper or restaurant owner, is one of the common perceptions of Britain’s ethnic minorities. But the evidence shows that there is huge diversity in rates of entrepreneurship between different ethnic minority groups, and that patterns of ethnic self-employment have been changing over time. Asian groups (Indian, Bangladeshi, Pakistani and Chinese) typically have higher rates of self-employment than Whites, while Black groups (Black African and Black Caribbean) have lower rates.

Since 1991, there has been some convergence in self-employment rates; some of those groups with the highest rates have seen a reduction in entrepreneurial activity while some of those with the lowest rates have seen their rates rise. In particular, Indian and Chinese males are now less likely to become self-employed than in the past.

Self-employment is generally higher than average among Asian groups: for example, while around 17% of White men in employment worked for themselves in 2001, this rate was 19% for Bangladeshis, 21% for Indians, 27% for Pakistanis and 28% for the Chinese. But among Britain’s other main ethnic groups, rates are much lower: Black Africans and Black Caribbean men had rates of around 13%.

There is also ethnic diversity in changes over time. Against a backdrop of a generally stable aggregate self-employment rate (after the dramatic increases of the 1980s), the period since 1991 has seen rates fall significantly for Indians (by around 2 percentage points) and the Chinese (6 percentage points). By contrast, Black Caribbean rates grew from 9% to 13% and Black African rates grew by around 1 percentage point. Only the Pakistani and Bangladeshi groups saw their rates stay broadly the same.

The researchers use data from the 1991 and 2001 Censuses, along with the Labour Force Survey, to analyse these trends. They find that much of the falling self-employment among Indians and Chinese can be ascribed to factors consistent with the greater assimilation of these groups into the UK labour market and education system.

In particular, the proportion of Chinese men with higher qualifications grew from 27% to 43% between 1991 and 2006 while for Indians the figures were 24% in 1991 and 40% in 2006. Formal educational qualifications are more highly rewarded in the paid employment sector so this growth in human capital has contributed to the decline in self-employment for these groups.

Similarly, the fact that the average working age Asian is now younger and more likely to be born in Britain, compared with 1991 also tends to reduce the propensity of entrepreneurship as older workers and first-generation immigrants have an increased likelihood of being in business for themselves.

One puzzle that emerges from the research is the fact that self-employment rates of Pakistani and Bangladeshi males have remained broadly stable in spite of the fact that this group shares many of the demographic trends of the other Asian groups. It is estimated that the Pakistani/Bangladeshi self-employment rate should have fallen by around 3 percentage points on the basis of demographic changes alone.

The fact that it did not suggests other factors are at work. It seems likely that discrimination in paid employment against Pakistani and Bangladeshi men is responsible for pushing them into self-employment, working long hours in relatively poorly rewarded sectors such as catering and taxi-driving. Given this, government policy should pay attention to the quality, as well as the quantity, of self-employment among all ethnic groups.

Notes for editors: ˜Changing Patterns of Ethnic Minority Self-Employment in Britain: Evidence from Census Microdata by Ken Clark and Stephen Drinkwater was presented at the Royal Economic Society’s 2007 annual conference at the University of Warwick, 11-13 April.

Stephen Drinkwater is at the University of Surrey; Ken Clark is at the University of Manchester. Their research was funded by the Joseph Rowntree Foundation.

For further information: Romesh Vaitilingam on 07768-661095 (email: romesh@compuserve.com).

Find more papers by Ken Clark and Stephen Drinkwater at EconPapers and search for more Internet resources on the issue of entrepreneurship at Intute: Economics.

Listen to this interview

Download audio file (kneller128.mp3)

Changes in exchange rates have little impact on UK manufacturing exports and are likely to have only a modest effect in reducing the countrys record trade deficit, according to research by Dr Richard Kneller and colleagues, presented to the Royal Economic Society’s 2007 annual conference at the University of Warwick.

Dr Kneller said:

The findings may surprise many people intuitively you would expect a strong pound to be bad for exports and a weak pound to lead to much greater exports.

But this research shows a different picture. It means that those concerned about the size of the trade deficit should not see a devaluation of sterling as a magic bullet solution to closing the gap.

In the most comprehensive research of its kind carried out in the UK, the researchers analysed exchange rate movements and export patterns of over 23,000 UK manufacturing firms over a 17-year period from 1987 to 2004. They find that changes in exchange rates:

• make little difference to a firms decision to start or stop exporting;
• make no difference at all to the level of exports of multinational firms;
• and while making some difference to domestic UK firms, they have a modest impact on exports.

Dr. Kneller said:

Our research shows that a drop in the value of the pound will not suddenly persuade British manufacturers to get out their foreign phrase books and start trying to sell overseas.

He said the analysis also showed that changes in exchange rates make no difference to the level of exports of multinationals based in the UK:

It would appear that multinationals are better able to internalise and offset currency risks. In the last few years, there has been a huge amount of foreign direct investment in the UK, which means that multinationals now account for at least a third of total UK exports.

But the report shows that exchange rates do have some effect on individual domestic UK manufacturers. For every 1% increase in an exchange rate index, a firms exports will drop by 1.28%. Usually exchange rate indices change by between 3 and 10 index points in a year.

But Dr. Kneller said

You have to put this into context. On average, exports account for just 5.6% of a domestic UK manufacturers business so, on average, a 10 point change in the exchange rate index will make about half a percent difference to total sales for a firm.

Notes for editors: Exchange Rates and Exports: Evidence from Manufacturing Firms in the UK by David Greenaway, Richard Kneller and Xufei Zhang was presented at the Royal Economic Societys 2007 Annual Conference at the University of Warwick, 11-13 April.

The authors are at the Globalisation and Economic Policy Centre at the University of Nottingham.

The latest balance of trade figures from the Office of National Statistics show that in December 2006, the UK imported £7.1bn worth of goods more than it exported. Last year, the UKs deficit on goods and services rose to a record £55.8 billion for the year as a whole, compared with a deficit of £44.6 billion in the previous year.

For further information: contact Romesh Vaitilingam on 07768-661095 (email: romesh@compuserve.com).

Find more papers by Richard Kneller at EconPapers and search Intute: Social Sciences for more Internet resources on exchange rates, exports and manufacturing. or get more economics updates at the Intute: Social Sciences blog.

Download audio file (res2.mp3)

Britain’s national minimum wage has not only raised the pay of low-paid employees. It has also led to a reduction in the rate of these employees absence through sickness  and hence improved their productivity. That is the finding of new research by Marco Ercolani and Martin Robson, presented at the Royal Economic Society Conference 2007. This effect might help to explain why the introduction of the minimum wage was greeted with apparent equanimity by many employers.

Employee sickness absence is widely recognised as a major problem for the economy. In recent years, for example, it has been calculated that the direct costs of sickness absence in terms of the value of lost output amount to over £11 billion per annum; around 1% of the country’ annual GDP.

On top of this, there are the indirect costs such as the loss of employee morale among those required to cover for absent colleagues. While many instances of employee sickness absence represent genuine cases of ill health, a significant proportion almost certainly does not.

The problem is that the latter cannot easily be distinguished from the former. So what can be done to discourage the shirkers and encourage them to turn up for work? This is an issue with which employers and governments have wrestled for a long time.

Studies by economists have demonstrated that there tends to be a negative relationship between employees pay and their rate of sickness absence: low-paid employees are more likely to be absent than those on higher earnings. So does this mean that paying higher wages will help to reduce employee sickness absence?

Unfortunately, the answer to this question is not straightforward. The negative relationship between employee pay and sickness absence might be observed because those with a poor record of absenteeism are less productive and are less likely to be promoted.

To answer the question properly, what is needed is the ability to study the effect of some kind of external shock that has the effect of raising an employees pay regardless of his or her level of productivity. Such an opportunity is provided by the introduction of the national minimum wage in April 1999.

The introduction of the minimum wage at a rate of £3.60 per hour for adults and £3.00 per hour for employees aged 18-2 is estimated to have raised the pay of around 1.2 million low-paid employees. Using data from the nationally representative Labour Force Survey, this study examines the effect of the change in the real value of employee pay induced by the introduction of the minimum wage on the rate of employee sickness absence.

The research finds that after controlling for the potential two-way relationship between sickness absence and earnings, a 1 percentage point increase in the rate of growth of the real value of employee earnings reduces the rate of sickness absence by about 0.05 percentage points, on average.

The introduction of the minimum wage is estimated to have raised the pay of those low-paid employees who were directly affected by around 4%. Thus, the introduction of the minimum wage is estimated to have led to a reduction in the rate of employee sickness absence of around 0.2 percentage points, on average a little under one-tenth of the average rate of sickness absence over the period studied.

By reducing the rate of employee sickness absence, the introduction of the minimum wage can therefore be seen to have had a beneficial effect on the productivity of low-paid employees, an effect that has not previously been highlighted in studies of the impact of minimum wage legislation. This effect might help to explain why the introduction of the minimum wage was greeted with apparent equanimity by many employers.

Notes for editors: Does Raising the Pay of Low-Wage Employees Reduce their Rate of Sickness Absence? Evidence from the Impact of Minimum Wage Legislation by Marco Ercolani and Martin Robson was presented at the Royal Economic Society’s 2007 annual conference at the University of Warwick, 11-13 April.

Marco Ercolani is at the University of Birmingham. Martin Robson is at Durham University.

For further information: contact Romesh Vaitilingam on 07768-661095 (email: romesh@compuserve.com)

Find more Internet resources on the issue of the Minimum Wage by searching Intute: Social Sciences and find more papers by Martin Robson and Marco Ercolani at EconPapers