3227_issue 8 2008:issue 2 2005

Copyright 2008 by theOxford Institute of Ageing AGEING HORIZONSIssue No. 8, 4–12 Age and Productivity Capacity: Descriptions, Causes and PolicyOptions Vegard Skirbekk, International Institute for Applied Systems Analysis
Abstract
understand which policies individuals and governmentscould undertake to increase the retirement age. This article reviews how work performance differs over thelife cycle by describing and discussing findings from variousapproaches. This includes managers’ evaluations, the quan- The Shape of the Age-Productivity Curve
tity and quality of goods and services produced by workersof different ages, the performance of age-mixed teams, to Identifying exceptional achievement at early or late ages is what extent the age distribution of employees depends on not difficult. For example, Goethe published the second part the type of work and how the age distribution changes due of Faust only in 1832 when he was 82 years old, while James to technological change and business cycle shocks, analy- Watson discovered the DNA molecule in 1953 at the age of ses of employer-employee datasets, descriptions of 25 with a few remarkable findings afterwards. Classic texts age-earnings profiles in settings where they could reflect in the age-performance literature, such as Crichton-Browne performance and the output of researchers and artists over (1905) and Osler (1906), are biased as they focus on excep- the life cycle. The causes of productivity variation by age tional performance by either senior or younger individuals to are also considered, with a particular focus on experience support their hypotheses of either an early or late productiv- and cognitive abilities. The findings suggest that produc- ity peak. Moreover, they only look at age performance among tivity tend to increase during the initial years in the labourmarket before it stabilizes and often declines towards the a small creative elite and use this to infer age variation in end of the working life. Productivity reductions at older performance for the population as a whole. ages are strongest in job tasks where problem solving, learn-ing and speed are important, while for work tasks where Classic studies of elite achievement by researchers, inno- experience and verbal abilities matter more, there is less or vators and artists by age tend to suggest that productivity no reduction in productivity among elderly workers. Trends peaks in the 30s and 40s, with substantially lower output in the age-productivity relation are discussed in relation to at younger and older ages, also when quality (e.g., journal changing work tasks and job requirements, combined with article citations) is controlled for (Lehman, 1953; Cole, changes in the requirement of skills (decline in demand for 1979; Simonton, 1988; Miller, 1999; Kanazawa, 2003; physical strength, increase in the need to learn new skills).
Oster and Hamermesh, 1998). Further evidence suggesting Policies that could be considered to raise productivity among that there is a negative association between scientific output senior workers include on-the-job training, education andpromotion of health. However, a later retirement could also and age from relatively early in adult life is given in studies raise incentives to update one’s own skills and work harder of economists and scientists by Bayer and Dutton (1977) at older ages (which may be achieved through pension and Bratsberg et al. (2003). However, a relatively flat reforms and wage liberalisation). Moreover, a better age- output profile for individuals aged 30–59 for 19th century mix in the workplace, allowing older and younger scientists is found by Dennis (1956), and for a limited individuals to benefit from their comparative advantages, sample of 20th century researchers by McDowell (1982).
is likely to improve overall productivity in ageing nations. Nobel Prize Laureates do their most important contributions Introduction
in their 30s, according to a study by Jones (2005). The onset Extending the working life is a popular response to popu- of the most innovate age phase, however, increased by about lation ageing in order to maintain economic prosperity and 6 years over the 20th century (which he suggests could be due sustainability. To achieve this, various policies are being to a longer training period as the knowledge base has considered, including changes in age-specific pension enti- expanded). The decline in innovative output, however, is tlements and age-discrimination laws (in particular, found to be age specific and constant which implies that the repealing laws for dismissing older workers without justi- plateau of peak performance has narrowed over time. fication). However, attempts to extend the age at retirementwill only be effective to the extent that productive employ- Entrepreneurial activity, start-up of new firms or expansions ment can be extended. This is important in order to ensure of existing ones, are more likely to be carried out by rela- that increased retirement age does not simply translate into tively young adults, according to findings from the 34 periods of unemployment. In this context, taking age- countries surveyed in the Global Entrepreneurship Monitor specific productivity into account is necessary in order to (GEMConsortium, 2004). Peak entrepreneurial activity is AGEING HORIZONS Issue No 8
OXFORD INSTITUTE OF AGEING
found among individuals aged 25–44. In support of a young Czaja and Sharit (1993) investigate the extent to which age entrepreneurial peak, Aubert et al. (2007) find, in a French has an impact on computer-based work performance and study, that the more innovative a firm is, the higher the find that increased age was associated with longer response wages of younger employees are relative to wages of other times and a greater number of errors for all tasks consid- ered. These task-quality/speed tests are potentially moreobjective as they do not rely on subjective assessment.
In order to describe age variation in performance of the However, they may be biased from the fact that the workers average worker rather than elite groups, one needs a differ- are selected in terms of age groups and occupational types ent set of productivity measures. Supervisors’ ratings are (Rubin and Perloff, 1993). Further, the time-limit common often used to identify the relation between the employee’s in such studies may bias results. For example older employ- age and his or her productivity. McEvoy and Cascio (1989) ees may maintain a higher work speed in the short period review 96 studies on the impact of the employee’s age on they are studied than what they would be able to do in a supervisors’ assessment and sales records and find no clear normal job situation (Salthouse and Maurer, 1996).
effect of age on productivity. Likewise, reviews by Warr(1994) and Waldman and Avolio (1986), based primarily A recent estimation approach is based on analyses of on supervisor assessment, find no or a slightly negative employer-employee datasets (e.g., Gelderblom and de impact of age on job performance. However, Remery et al.
Koning, 2002; Hægeland and Klette, 1999). A key strength (2003) find that older individuals are seen as less produc- of these estimates are the large samples, which can encom- tive in particular in firms with a higher proportion of senior pass most workers in some of the main industries in a workers, which is where knowledge about older individu- country; several samples include a few million individuals als’ work capacities is likely to be highest, in a survey 1007 and thousands of firms. The most common finding from these studies is a hump-shaped relation between job perform-ance and age. An overview over analyses of the effect of A general problem with most approaches used to measure age on productivity using employer-employee datasets are age variation among workers is the sample selection given in the Annex. Of the 14 studies considered, 11 find problem. Good workers get promoted, while inefficient a productivity decline in the 50s relative to the 30s and 40s, workers may lose their jobs. Hence, positive selection can two have inconsistent results, while one finds that produc- increase by age, which could lead to bias in the estimates tivity peaks among the oldest workers. However, bias may of the oldest age groups’ working capacity. Self-assessed come from the fact that many determinants of a firm’s “work ability” is sometimes considered, and estimates from value-added, such as capital levels, are either omitted or Finland suggest a clear decline by age of workers (van poorly measured. The reverse causality problem could also Ours et al. 2007). Solem (2006) presents evidence from be an issue – a company’s success can increase the number Norwegian surveys on subjective general work ability that of new employees and lead to a younger age structure, shows a decline from the 30s to the 60s. Nevertheless, which could mean that a young age structure could be the respondents believe that they are equally capable of perform- consequence rather than the cause of a company’s success.
ing their work over the same age interval; although generalwork ability declines with age. This may be because rele- Wages are often determined by other factors than individual vant abilities are fully maintained and the worker is contribution to the firms’ value-added, including the role of increasingly well matched to the type of work task he/she unions, uncertainty about new workers productivity and performs. A particular problem with managers’ ratings and delayed payment contracts where performance while being self-evaluations is that these evaluations can be biased as young results in higher earnings for those still employed while they are subjective. Evaluations of older workers could they are older (Agell and Lundborg, 1995; Freeman, 1982; therefore be inflated due to loyalty concerns and as a reward Hutchens, 1989). Age-earnings profiles can, however, for past achievement – or in the case of self-reports, as a provide information on productivity differences in settings means of self-justification, or even self-degradation.
where wages are likely to reflect actual productivity. One Discriminatory attitudes among managers towards older or example is a study by Lazear and Moore (1984) who examine younger individuals can also affect an age-productivity esti- the difference between earnings profiles of the self-employed mate, which reduces the validity of this approach (Levy, and salary workers. Kotlikoff and Gokhale (1992) study age- variation in the wages offered when firms hire newemployees. Their findings suggest managers have earliest Measuring the impact of age on job performance is some- productivity maximum, while office and sales-workers have times based on measures of the quantity and quality of a somewhat older peaks, though they all reach their highest worker’s output. Studies based on this approach tend to find productivity in the 40s, with sharp declines thereafter. that older employees have lower productivity levels. Astudy of a several industries from the U.S. Department of A study by Boot (1995) describes age-earnings profiles for Labor (1957) finds that job performance increases until the British workers in the first half of the 19th century, when age of 35 and steadily declines thereafter. At the end of the there were few regulations in the labour market. For the career, productivity declines by 14% in the men’s footwear physically demanding work analyzed here, men reached industry, and 17% in the household furniture industry.
their peak earnings in their early 30s, and wages decreased AGEING HORIZONS Issue No 8
OXFORD INSTITUTE OF AGEING
substantially from around 40 years of age. In a similar suggest that older workers have a modest negative impact on study, Johnson (2003) looks at British manual workers earn- productivity growth, while Nishimura et al. (2002) find that ings from the 1830s to the 1930s, and finds a stable the share of well-educated workers above 40 years raised tech- age-earnings pattern where wages reach their peak in the nological progress in the 1980s, but decreased it in the 1990s mid-30s and remain stable or decline slightly thereafter. when investigating technical progress and growth in Japan1980–1998. It is not clear why certain relations between the Labour force attachment trends can also reveal information age structure and economic growth levels exist. They could be about age differences in productivity. If older workers cope due to other factors than age variation in work performance, less well with changes in the workplace, then rapid changes such as high savings levels among certain age groups (i.e., should affect them worse than younger age groups. Bartel higher capital levels could raise growth) or omitted factors in and Sicherman (1993) put forward evidence that the risk of the regressions (which both affect economic growth as well as job loss is in fact greater among older workers when the mortality, fertility or migration levels).
rate of technological change is highest. This finding is alsosupported in studies based on inter-sectoral data (Ahituv and Causes of Age Variation in Work
Zeira, 2000). Technological improvements may also dispro- Performance Relation
portionally benefit the young, and in a study of Germanfirms, Hujer and Radic (2005) find that older highly-skilled Workers’ productivity levels differ by age for many reasons, workers benefit less from technological improvements including the length of work experience, cognitive func- tioning, education, physical abilities, stamina, health, familyand care obligations, motivation, energy, matching of the Age diversity has been suggested to have a positive effect on worker to the task, loyalty and personality. productivity through better “age-matching” between sales-persons and customers and through complementarity of Average muscle strength decreases by roughly 10% per different age groups, but the evidence seems inconclusive.
decade for ages 20–60, by approximately 15% each decade Experimental research suggests cooperation is highest when for ages 60–80, and 30% each decade after age 80 (Mazzeo, juniors and seniors are mixed together (Charness and Villeval, 2000). De Zwart et al. (1995) show that aerobic capacity 2006). However, Leonard and Levine (2003) considered 800 peaks somewhere in the 20s and declines by around 1% per retail stores employing 70,000 individuals and found that age year. Flexibility decreases with age, which makes it diffi- diversity was negatively related to sales. Hamilton et al. cult to adopt certain working positions (Bosek et al., 2005). (2004) studied the productivity of workers in a garment plantthat had shifted from individual piece rate payments to team Moreover, cognitive skills have been shown to be increas- piece rate payments, and again found that greater diversity ingly important to work tasks over time, as both the work in age was associated with lower productivity gains. A intensity and the industrial composition change over time, European study on age diversity suggests that while in a while physical strength is becoming decreasingly important German car assembly plant, age diversity increases the prob- (Broadberry, 1997; Spitz-Oener, 2006; Skirbekk, 2008).
ability of calling in sick, evidence from Finnish firms show Cognitive ability test scores have been found to be more a positive effect of age diversity (van Ours et al. 2007). Van closely correlated with labour market performance than any Ours et al. suggest that this seeming contradiction is due to other observable characteristics among job candidates different levels of observation, where age diversity at the team level (German case) may be negative (as age diversitymay imply impeded communication), while diversity at the Some cognitive abilities decline with age, while others tend plant level (Finnish case) may mean that in some parts of the to be relatively robust over the life cycle. A division can plant (e.g., management) there are many older (wise, expe- be drawn between crystallised abilities, which remain at a rienced) workers while in other parts (e.g., production), there high functional level until late in life, and fluid abilities, are many younger (fit, flexible) workers. Hence, comple- mental abilities that are strongly reduced over the life span mentarity at the firm level may well be consistent with the (Horn and Cattell, 1966; Verhaegen and Salthouse, 1997).
Crystallised abilities depend on accumulated knowledge,and includes semantic meaning and vocabulary size. The Several macroeconomic investigations find a relation between second group, fluid abilities such as perceptual speed and national/regional age structure and economic growth.
reasoning abilities, tend to relate to performance and speed However, results diverge and while Brunow and Hirte (2006) of solving tasks related to new materials. find that 30–44 year olds boost per capita output growth acrossEuropean regions, Feyrer (2005) finds that the 40–49 year old Schwartzman et al. (1987) find that verbal skills (crys- age group is the strongest associated with productivity growth tallised abilities) remain virtually unchanged at older ages, using a large panel of developed and developing countries, while reasoning and speed (fluid abilities) decrease from while Lindh and Malmberg (1999) and Malmberg and Lindh early adulthood, based on psychometric test results of men (2002) find that the share of 50–64 year olds is associated with in different age groups. In a test-retest study of twins, Blum economic growth in the OECD from 1850–1990. Moreover, et al. (1970) provide similar findings: vocabulary size is Tang and MacLeod (2006), using Canadian regional data, observed to remain constant from young to old ages despite AGEING HORIZONS Issue No 8
OXFORD INSTITUTE OF AGEING
a general reduction in other cognitive abilities. A decline nity of part-time work increases the opportunities for active over the working life span in some mental abilities has labour force participation among the elderly. These types of been found to be similar for both men and women, and the flexible work settings contrast the situation for assembly line same patterns are found across different countries and workers in the past, where constant and continuous attention cultures (Maitland et al., 2000; Park et al., 1999).
Furthermore, individuals with high ability levels are subjectto the similar changes in cognitive functioning as those Improving Productivity and Employment at
with low ability levels (Baltes and Mayer, 1999; Deary et Older Ages
al., 2000). Age-related reductions in memory and learning Romans sometimes described old age as mala aetas, the bad capabilities have been documented among many non-human age, and young age as bona aetas, the good age (Parkin, species, ranging from fruit flies to primates (Bunk, 2000; 2003). However, age-related changes to physique and mind have changed since Roman times, and cognitive and healthdeterioration takes place at increasingly later ages in contem- In some occupations, the cognitive abilities that remain porary societies. And surroundings have changed, where the stable are the ones most closely correlated to job success.
introduction of technologies compensating declines in Senior employees can remain highly productive within a sensory and muscular capabilities, electric wheelchairs, field that they know well and where relatively long expe- more user-friendly computer systems, effective medicines rience is beneficial. Tacit knowledge, procedural knowledge against pain and hypertension and working at home oppor- used to solve everyday problems, tends not to decrease at tunities allow one to continue to function professionally in older ages. The age-robustness of this ability could explain older ages in a large number of work places. why many older managers perform as good as youngerones (Colonia-Willner, 1998). However, when performing There is evidence that mental abilities have improved over unfamiliar work, workers have to rely on the ability to the course of the last century (Tuddenham, 1948; Flynn, learn and to adjust exactly those skills that decline most with 1987; Neisser, 1997). Improvement in cognitive skills have age. Senior individuals are less able than young individu- been found to also take place for individuals aged 50–79 als to reorient themselves to new task requirements and to during the last decades (Romeu Gordo, 2005) which is solve novel problems (Smith, 1996) and age-induced produc- likely to be related to rapid growth in education and on- tivity reductions may increase with the complexity of new the-job training (Lutz et al., 2007; Riphahn and work tasks (Myerson et al., 1990). Hence, staying within Trübswetter, 2006). Individuals from younger cohorts are one field of expertise relates to high productivity, while likely to be more trained and motivated when taking ability changing between fields can lower productivity much more tests, as such tests are increasingly being used in job candi- (e.g., Cole, 1979; Rybash et al., 1986). date selection processes (Jenkins, 2001). Experience is often quoted by employers as one of the most Paralleling the sustained decline in mortality (which has important determinants of job performance (Bellmann, 2007; taken place for at least two centuries in Northern Europe), Golini, 2004). Salthouse (1984) uses typists as an example disability-free life expectancies have increased (Crimmins of a profession where experience alleviates the impact of et al., 1997; Manton et al., 1997; Lee, 2003; Romeu cognitive reductions. He finds that older typists compen- Gordo, 2005; Schoeni et al., 2001). The morbidity decrease sate for their reduced speed through the use of more efficient is partly related to less physically demanding work tasks.
work strategies compared to their younger counterparts.
Costa et al. (1986) argues that 29% of the observed decline Ericsson and Lehmann (1996) argue that it takes roughly in chronic disease rates and 75% of the decline in back prob- 10 years to achieve expert competence in games and situ- lems in the 20th century was caused by a shift to non-manual ations where strategic and analytic competence is important, labour and a reduction in the physical load. Not only has such as chess. The 10-year estimate is also supported by work become less strenuous, but estimates put forward by findings from a variety of job domains, ranging from live- Ausubel and Grubler (1995) suggest that in France, stock evaluation and X-ray analysis to scientific performance Germany, the UK, USA and Japan, the number of hours in medical and natural sciences (Phelps and Shanteau, 1978; worked per year declined by at least a third from 1850 to Lehman, 1953; Lesgold, 1984; Raskin, 1936). 1987, which is likely to imply a substantial reduction in thehealth requirements to work.
A Norwegian survey (Econ, 1998) shows that only 0.6% ofthe employers preferred workers with a length of experience To extend the productive working life at older ages, a above 10 years. As the average age of entering the labour government could follow many different policies. Taxes, force in the OECD is in the early 20s (depending on educa- earnings and pension systems need to be restructured in tional levels, OECD, 1999), this would suggest that 99.4% order to increase incentives for older individuals to continue of the employers find workers sufficiently experienced in the to work rather than to retire. However, the question is not 30s, which would exclude experience as a major advantage only about creating financial incentives for a long and between workers in the 50s and 60s compared to workers in productive working life, but also about the need to alter the 30s and 40s, although all have an advantage compared to norms, beliefs and behaviour that affect labour force partic- workers in their 20s. The increased prevalence and opportu- AGEING HORIZONS Issue No 8
OXFORD INSTITUTE OF AGEING
Many elderly would like to continue working when they reach proportion of the age group remains active. Socialising has retirement age, while employers are often hesitant to employ been shown to be good for mental health and for many, them, making it difficult to increase retirement ages. High particularly men, the work place is the key place to meet costs are often seen as the main disadvantage of older workers others (Cohen, 2004; Melchior et al., 2003).
(Golini, 2004; Munnell et al., 2006). Pension payments andhealth insurance often increase by age, and as wages may not As the current population is likely to be urban, have seden- be adjusted for this, it implies that older individuals are cost- tary occupations, and use motorized transport, exercise lier for firms (Scott et al. 1995). In firms with steep wage should be encouraged through increased availability of green profiles and pension benefits, employment opportunities have space and sports facilities. More exercise can increase work been found to be worse for elderly in the US (Hirsch et al., capacity at older ages. James and Coyle (1998) find that 2000), Hong Kong (Heywood et al., 1999) and the UK regular exercise improves the working memory function among older men, while obesity has been found to havenegative effects on cognitive performance net of education, For example, in Japan, it is customary for those who retire occupation, cigarette smoking, alcohol consumption, total from a normal working career to enter a lower-wage but cholesterol, and a diagnosis of diabetes (Elias et al., 2003; relatively prestigious second career in their 60s (Clark et Geroldi et al., 2003; Gustafson et al., 2003). al., 2006). Ichino et al. (2006) find that when displacedolder workers’ minimum wages they are willing to accept Encouraging new nutritional habits and increased exercise fall faster than those of prime-age workers, they catch up could lower the negative impacts on productivity, such as in terms of employment. A comparison of earnings levels absenteeism, disability, workplace injuries and health costs and employment rates across the OECD suggest that wage (Schmier et al., 2006; Colditz, 1999). Cereals and fish, red flexibility and willingness to accept lower wages relates to wine or even aspirin could possibly strengthen cognitive functioning (Nilsson et al., 2003; Solfrizzi et al., 2003).
Governments could also impose stronger non-smoking legis- However employers/governments also need to make funda- lation, as smoking has been found to negatively affect mental changes to how they view older workers. Flexible cognitive functioning, also when adjusting for the negative retirement and part-time positions, improved career planning selection of smokers (Deary et al., 2003). Moreover, better and more internal mobility, preventive measures and occu- system design can reduce computer problems for older pational health programmes could increase work productivity of the elderly (see Lindley et al., 2006 for “good practice”examples). Age discrimination must end, job training oppor- In spite of the seemingly unavoidable age-related reductions tunities should be given without age constraints, and in cognitive abilities, targeted training programmes and age-specific dismissal legislation should be phased out. One engagement in complex work tasks may provide a way of could, if possible, reorganize to allow the elderly to perform halting the decline. Schooler et al. (1999) find that the degree work tasks where they have their comparative advantages, of job complexity is associated with the level of mental func- which is where experience, management and communication tioning at older ages, which could be due to continued learning skills are important rather than tasks which require high in the workplace. Schaie and Willis (1986) conclude that train- processing speed and rapid reorientation.
ing programmes can stabilise, or even reverse, age-relateddeclines in inductive reasoning and spatial orientation among More pensioners means that taxes need to be increased to many individuals. Similar evidence is presented by Ball et al.
support more pensioners, and higher taxes slow the creation (2002), who find that persons who exercise the use of indi- of jobs. However, a change of norms is necessary, where vidual abilities such as speed, reasoning and memory, enhance a higher pension age should become the custom, and seen the functional level of these abilities. Katzman (1993) argues as necessary and fair. Further, beliefs that justify early that participating in educational courses increases synaptic retirement, e.g., that retiring early increases job opportu- density in the neocortical association cortex, and could there- nities for the young, should be abandoned. In countries fore delay the onset of dementia by up to 4–5 years. Using a such as the Netherlands, France, Germany, Spain, Italy British industry dataset from 1983–1996, Dearden et al. and the UK, between 40% and 70% of the individuals agree (2006) find that training is associated with productivity that “older people should be forced to retire when jobs are increases, and more so than what wage increases would scarce” (van Ours et al., 2007). suggest. Likewise, in separate studies of German firm panel-datasets, Zwick (2002) and Kuckulenz and Maier (2006) If one expects a later retirement, one’s training investment identify large productivity increases in response to the train- decisions are likely to change. A longer remaining working life implies that it will be optimal for individuals to investmore in updating skills and to increase work efforts at older Conclusion
ages. When the elderly expect to work longer, they will haveincentives to invest more in training and to work harder.
The evidence put forward in this article suggests that job They will also continue to update, maintain and extend performance often does decrease at older ages, but not their professional networks, which will be easier if a larger for all work tasks and less so in occupations where abil- AGEING HORIZONS Issue No 8
OXFORD INSTITUTE OF AGEING
ities that do not decline by age are important. Moreover, the elderly to benefit from their comparative advantages the elderly’s productive potential is likely to have and effectively extend the working life. Changes in increase over time as modern jobs decreasingly rely on incentive systems, regulations and habits that lead to less strength, cognitive abilities and health of the elderly smoking, better nutrition and more exercise could further improve, and part-time flexible work arrangements have decrease disability levels and raise cognitive functioning become commonplace. More training, better work organ- is likely to help increase work performance levels and isation, and more flexible earnings systems could allow labour force participation at older ages. References
Agell, J. and P. Lundborg (1995) Theories of pay and unemployment: Transfers, and the Macroeconomy, Tokyo, Japan.
Survey evidence from Swedish manufacturing firms. Scandinavian Cohen, S. (2004) Social relationships and health. American Journal of Economics, 97, pp. 295–307.
Psychologist, November, pp. 676–684.
Ahituv, A. and J. Zeira (2000) Technical progress and early Colditz, G. A. (1999) Economic costs of obesity and inactivity. Med. retirement. CEPR Discussion Paper No. 2614. London: Centre for Sci. Sports Exerc., 31(11), pp. 663–S667.
Cole, S. (1979) Age and scientific performance. American Journal of Aubert, P. and B. Crépon (2004) La productivité des salariés âgés: une Sociology, 84(4), pp. 958–977.
tentative d’estimation [Productivity of older workers: a tentative Colonia-Willner, R. (1998) Practical intelligence at work: Relationship estimate]. Économie et statistique, 368, pp. 95–119.
between aging and cognitive efficiency among managers in a bank Aubert, P., E. Caroli, and M. Roger (2007) New technologies, environment. Psychology and Aging, 13(1), pp. 45–57.
organisation and age: Firm level evidence. The Economic Journal, Costa, P. C., R. R. McRae, A. B. Zonderman, H. E. Barbano, B.
Lebowitz, and D. M. Larson. (1986) Cross-sectional studies of Ausubel, J. H. and A. Gruebler (1995) Working less and living longer: personality in a national sample: Stability in neuroticism, Long-term trends in working time and time budgets. Technological extraversion and openness. Psychology and Aging, 1, pp. 144–149.
Forecasting and Social Change, 50 (3), pp. 113–131.
Crépon, B., N. Deniau, and S. Perez-Duarte (2002) Wages, Ball, K., D. B. Berch, K. F. Helmers, J. B. Jobe, M. D. Leveck, M.
productivity and worker characteristics. A French perspective.
Marsiske, J. N. Morris, G.W. Rebok, D. M. Smith, S. L.
Mimeo. Paris: National Institute for Statistics and Economic Studies Tennstedt, F.W. Unverzagt, and S. L.Willis (2002) Effects of cognitive training interventions with older adults. Journal of the Crichton-Browne, J., (1905) The Prevention of Senility and a Sanitary American Medical Association, 288(18), pp. 2271–2281.
Baltes, P. and K. U. Mayer (eds.) (1999) The Berlin aging study: Crimmins, E. M., Y. Saito, and S. L. Reynolds (1997) Further Aging from 70 to 100. Cambridge: Cambridge University Press.
evidence on recent trends in the prevalence and incidence of Barrington, L. and K. Troske (2001) Workforce diversity and disability among older Americans from two sources: The LSOA and productivity: An analysis of employer-employee matched data. New the NHIS. Journal of Gerontology, 52B(2), pp. 59–71.
York: Conference Board Economics Program Working Paper Series Czaja, S. and C. C. Lee (2007) The impact of aging on access to technology. Universal Access in the Information Society, 5(4), pp.
Bartel, A. P. and N. Sicherman (1993) Technological change and retirement decisions of old workers. Journal of Labor Economics, Czaja, S. and J. Sharit (1993) Age differences in the performance of computer-based work. Psychol. Aging, 8(1), pp. 59–67.
Bayer, A. E. and J. E. Dutton (1977) Career, age and research Daniel, K. and J. Heywood (2007) The determinants of hiring older professional activities of academic scientists. Journal of Higher workers: UK evidence. Labour Economics, 14, pp. 35–51.
Education, 48(3), pp. 252–282.
de Zwart, B., M. Frings-Dresen, and F. van Dijk. (1995) Physical Bellmann, L. (2007) Einstellungen Älterer aus betrieblicher Sicht.
workload and the ageing worker: A review of the literature.
[Firms’ attitudes towards elderly]. Paper presented at the Conference International Archives of Occupational and Environmental Health, “Chancen für Ältere? Neue Forschungsergebnisse zur Arbeitsmarktsituation Älterer” held on March 2, 2007 in Berlin.
Dearden, L., H. Reed, and J. Van Reenen (2006) The impact of Blum, J. E., L. F. Jarvik, and E. T. Clark (1970) Rate of change on training on productivity and wages: Evidence from British Panel selective tests of intelligence: A twenty-year longitudinal study of Data. Oxford Bulletin of Economics and Statistics, 68(4), pp.
aging. Journal of Gerontology, 25, pp. 171–176.
Boot, H. M. (1995) How skilled were Lancashire cotton factory Deary, I. J., A. Pattie, M. D. Taylor, M. C. Whiteman, J. M. Starr, workers in 1833? Economic History Review, XLVIII (2), pp.
and L. J. Whalley (2003) Smoking and cognitive change from age 11 to age 80. J. Neurol. Neurosurg. Psychiatry, 74, pp. 1006–1007.
Bosek, M., B. Grzegorzewski, A. Kowalczyk, and I. Lubinski (2005) Deary, I. J., L. J. Whalley, H. Lemmon, J. R. Crawford, and J. M.
Degradation of postural control system as a consequence of Parkinson’s Starr (2000) The stability of individual differences in mental ability disease and ageing. Neuroscience Letters, 376, pp. 215–20.
from childhood to old age. Follow-up of the 1932 Scottish mental Bratsberg, B., J .F. Ragan, Jr., and J. T. Warren (2003) Negative survey. Intelligence, 28(1), pp. 49–55.
returns to seniority: New evidence in academic markets. Industrial Dennis, W. (1956) Age and productivity among scientists. Science, 27, and Labor Relations Review, 56(2), pp. 306–323. Broadberry, S. N. (1997) Forging up, falling behind and catching up.
ECON (1998) Arbeidsmarkedspolitikken overfor eldre arbeidssøkere A sectoral analysis of Anglo-American productivity differences, [Labour Market Policies towards elderly job seekers]. Oslo, 1870–1990. Research in Economic History, 17, pp. 1–37.
Norway: Report for the Ministry of Local Government and Regional Brunow, S. and G. Hirte (2006) Age structure and regional growth.
Development. Econ Report No. 98/97. Jahrbuch für Regionalwissenschaft/Review of Regional Research, Elias, M. F., P. K. Elias, L. M. Sullivan, P. A. Wolf, and R. B.
d’Agostino (2003) Lower cognitive function in the presence of Bunk, S. (2000) Cognition and aging. A rare colony of old macaques obesity and hypertension: the Framingham heart study. International gives clues to age impairment. The Scientist, 14, pp. 18–19.
Journal of Obesity 27, pp.260–268.
Charness, G. and M.-C.Villeval (2006) Cooperation, competition, and Ericsson, K. A. and A. C. Lehmann (1996) Expert and exceptional risk attitudes: An intergenerational field and laboratory experiment.
performance: Evidence of maximal adaptation to task constraints.
Institute for the Study of Labour (IZA) Discussion Paper 2574.
Annual Review of Psychology, 47, pp. 273–305.
Clark, R., N. Ogawa, and R. Matsukura (2006) Population aging, Feyrer, J. (2005) Demographics and productivity. Draft. Hanover, Changing retirement policies, and lifetime earnings profiles. Paper New Hampshire: Dartmouth College, Department of Economics.
presented at the Conference on Population Aging, Intergenerational Flynn, J. R. (1987) Massive IQ gains in 14 nations. What IQ tests AGEING HORIZONS Issue No 8
OXFORD INSTITUTE OF AGEING
really measure. Psychological Bulletin, 101, pp. 171–191.
Paper 11359. Cambridge, MA: National Bureau of Economic Freeman, R. B. (1982) Union wage practices and wage dispersion within establishments. Industrial and Labor Relations Review, 36, Kanazawa, S. (2003) Why productivity fades with age: The crime-genius connection. Journal of Research in Personality, 37, pp. 257–272.
Gelderblom, A. and J. de Koning (2002) Exclusion of older workers, Katzman, R. (1993) Education and the prevalence of dementia and productivity and training. In: K. Schömann and P.J. O’Connell Alzheimer’s disease. Neurology, 43, pp. 13–20.
(eds.) Education, Training and Employment Dynamics. Cheltenham: Kotlikoff, K. and J. Gokhale (1992) Estimating a firm’s age- productivity profile using the present value of workers’ earnings. The GEMConsortium (2004) National team reports. Accessed 21 April 2005 Quarterly Journal of Economics, 107(4), pp. 1215–1242.
at http://www.gemconsortium.org/category_list.asp?cid=164 Kuckulenz, A. and M. Maier (2006) Heterogeneous returns to training.
Geroldi, C., L. Ferrucci, S. Bandinelli, et al. (2003) Mild cognitive An analysis with German data using local instrumental variables.
deterioration with subcortical features. Prevalence, clinical ZEW Discussion Paper No. 06–002. Mannheim, Germany: Centre characteristics, and association with cardiovascular risk factors in community-dwelling older persons. J. Am. Geriatr. Soc., 51, pp.
Lazear, E. P. and R. L. Moore (1984) Incentives, productivity, and labor contracts. The Quarterly Journal of Economics, 99(2), pp. 275–296.
Golini, A. (2004) Demographic trends, aging and labour: The work of the Lee, R. (2003) The demographic transition: Three centuries of young elderly: Is there willingness and possibility? Paper presented at fundamental change. Journal of Economic Perspectives, 17, pp.
the Vienna Demographic Colloquium on February 23, 2004.
Grund, C. and N. Westergård-Nielsen (2005) Age structure of the Lehman, H. C. (1953) Age and Achievement. Princeton, NJ: Princeton workforce and firm performance.” Working paper WP-05–10.
Aarhus, Denmark: Aarhus School of Business.
Leonard, J. S. and D. I. Levine (2003) Diversity, discrimination and Gustafson, D., E. Rothenberg, K. Blennow, et al. (2003) An 18-year performance. IIRUPS Working Paper 091–03. Berkeley, CA: follow-up of overweight and risk of Alzheimer disease. Arch Intern Institute of Industrial Relations, University of California, Berkeley.
Lesgold, A. (1984) Acquiring expertise. In: J. Anderson and S.
Hægeland, T. and T. J. Klette (1999) Do higher wages reflect higher Kosslyn (eds.) Tutorials in Learning and Memory: Essays in Honor productivity? Education, gender and experience premiums in a of Gordon Bower. New York: Freeman, pp. 31–60.
matched plant-worker data set. In: J. L. Haltiwanger, J. R. Spletzer, Levy, B. R. (2003) Mind matters: Cognitive and physical effects of J. Theeuwes, and K. Troske (eds.) The Creation and Analysis of aging self-stereotypes. Journal of Gerontology: Psychological Employer-Employee Matched Data. Amsterdam: North Holland, pp.
Sciences, 58B(4), pp. 203–211.
Lindh, T. and B. Malmberg (1999) Age structure and growth in the Haltivanger, J. C., J. I. Lane, and J. R. Spletzer (1999) Productivity OECD. Journal of Population Economics, 12(3), pp. 431–449.
differences across employers. The roles of employer size, age and Lindley, R., et al. (2006) Ageing and Employment: Identification of human capital. American Economic Review, 89 (2), pp. 94–98.
Good Practice to Increase Job Opportunities and Maintain Older Hamilton, B. H., J. A. Nickerson, and H. Owan (2004) Diversity and Workers in Employment. Report for the European Commission.
Productivity in Production Teams. St. Louis, Missouri: John M. Olin Brussels: European Commission, Directorate-General for School of Business, Washington University. Draft, May.
Employment, Social Affairs and Equal Opportunities, Warwick IER.
Hellerstein, J. and D. Neumark (2004) Production function and wage Lutz, W., A. Goujon, S. K.C., and W. Sanderson (2007) equation estimation with heterogeneous labor: Evidence from a new Reconstruction of populations by age, sex and level of educational matched employer-employee data set. NBER Working Paper 10325.
attainment for 120 countries for 1970–2000. In: The Vienna Cambridge, MA: National Bureau of Economic Research.
Yearbook of Population Research 2007. Vienna: Verlag der Hellerstein, J. K. and D. Neumark (1995) Are earnings profiles steeper Österreichischen Akademie der Wissenschaften (forthcoming).
than productivity profiles? Evidence from Israeli firm-level data.
Maitland, S. B., R. C. Intrieri, K. W. Schaie, and S. L. Willis (2000) Journal of Human Resources, XXX (1), pp. 89–112.
Gender differences and changes in cognitive abilities across the adult Hellerstein, J. K., D. Neumark, and K. R. Troske (1999) Wages, life span. Aging, Neuropsychology, and Cognition, 7(1), pp. 32–53.
productivity and worker characteristics: Evidence from plant level Malmberg, B. and T. Lindh (2002) Population change and economic production function and wage equations. Journal of Labor growth in the western world, 1850–1990. Paper presented at the 3rd Economics, 17, pp. 409–446.
Max Planck Institute Conference on Demographic Macroeconomic Heywood, J., L. Ho, and X. Wei (1999) The determinants of hiring Modeling, held in Rostock, Germany, September 2002.
older workers: Evidence from Hong Kong. Industrial and Labour Manton, K. G., L. Corder, and E. Stallard (1997) Chronic disability Relations Review, 52, pp. 444–459.
trends in elderly United States populations: 1982–1994. Proceedings Hirsch, B. T., D. MacPherson, and M. Hardy (2000) Occupational age of the National Academy of Science, 94, pp. 2593–2598.
structure and access for older workers. Industrial and Labour Mazzeo, R.S. (2000) Current comment: Exercise and the older adult.
Relations Review, 53, pp. 401–418.
American College of Sports Medicine. Available at ACSM on the Horn, J. L. and R. Cattell (1966) Refinement and test of the theory of fluid and crystallized intelligence. Journal of Educational Psychology www.acsm.org/health%2Bfitness/pdf/currentcomments/eoa.pdf McDowell, J. (1982) Obsolence of knowledge and career publication Hujer, R, and D. Radic (2005) Age and skill biased technological profiles: Some evidence of differences among fields in costs of change: A multiple treatment approach using a linked employer interrupted careers. American Economic Review, 72(4), pp. 752–768.
McEvoy, G. M. and W. F. Cascio (1989) Cumulative evidence of the doku.iab.de/veranstaltungen/2006/CAFE_2006_K1_Radic.pdf relationship between employee age and job performance. Journal of Hutchens, R. (1989) Seniority, wages and productivity: A turbulent Applied Psychology, 74(1), pp. 11–17.
decade. Journal of Economic Perspectives 3(4), pp. 49–64.
Melchior, M., L. F. Berkman, I. Niedhammer, M. Chea, and M.
Ichino, A., G. Schwerdt, R. Winter-Ebmer, and J. Zweimüller (2006) Too Goldberg (2003) Social relations and self-reported health: A Old to Work, Too Young to Retire? Paper presented at the Conference of prospective analysis of the French Gazel cohort. Social Science and European Association of Labour Economists. Available at: Medicine, 56, pp. 1817–1830.
http://www.eale.nl/conference2006/Papers%20Friday%2017.00%20–% Miller, G. F. (1999) Sexual selection for cultural display. In: R.
Dunbar, C. Knight, and C. Power (eds.) The Evolution of Culture: Ilmakunnas, P., M. Maliranta, and J. Vainiomäki (2004) The role of An Interdisciplinary View. New Brunswick: Rutgers University employer and employee characteristics for plant productivity. Journal of Productivity Analysis, 21, pp. 249–276.
Minois, N. and E. Le Bourg (1997) Hypergravity and aging in Jenkins, A. (2001) Companies’ use of psychometric testing and the drosophila melonagaster. Conditioned suppression and habituation of changing demand for skills: A review of the literature. Working the proboscis extension response. Aging, Experimental and Clinical Paper. London: Centre for the Economics of Education, London Munnell, A., S. Sass, and M. Soto (2006) Employers attitudes towards Johnson, P. (2003) Age, gender and wage in Britain 1830–1930. In: older workers: Survey results. Series 3, July. Boston College, Scholliers, P. and L. Schwarz (eds.) Experiencing Wages: Social and Massachusetts: Center for Retirement Research.
Cultural Aspects of Wage Forms in Europe since 1500. New York: Myerson, J., S. Hale, D. Wagstaff, L. W. Ponn, and G. A. Smith (1990) The information loss model: A mathematical theory of age- Jones, B. F. (2005) Age and great invention. NBER Working related slowing. Psychological Review, 97, pp. 475–486.
AGEING HORIZONS Issue No 8
OXFORD INSTITUTE OF AGEING
Neisser, U. (1997) Rising Scores on Intelligence Tests. American Schmidt, F. L. and J. E. Hunter (1998) The validity and utility of selection methods in personnel psychology: Practical and theoretical Nilsson, S. E., B. Johansson, S. Takkinen, S. Berg, S. Zarit, G.
implications of 85 years of research findings. Psychological Bulletin, McClearn, and A. Melander (2003) Does aspirin protect against Alzheimer’s dementia? A study in a Swedish population-based Schmier, J. K., M. L. Jones, and M. T. Halpern (2006) Cost of obesity in sample aged =80 years. European Journal of Clinical the workplace. Scand J Work Environ Health, 32 (1), pp. 5–11.
Pharmacology, 59(4), pp. 313–319.
Schneider, L. (2006) Sind ältere beschäftigte weniger produktiv? Eine Nishimura, K. G., K. Minetaki, M. Shirai, and F. Kurokawa (2002) empirische analyse anhand des LIAB. [Are older workers less produc- Effects of information technology and aging work force on labor tive? An analysis based on the LIAB]. Halle, Germany: Institut für demand and technological progress in Japanese industries: Wirtschaftsforschung Halle (IWH), Working Paper Nr. 13.
1980–1998. Discussion Paper CIRJE-F-145. Tokyo: Faculty of Schoeni, R. F., V. A. Freedman, and R. B. Wallace (2001) Persistent, consistent, widespread, and robust? Another look at recent trends in OECD (1999) Thematic Review of the Transition from Initial Education old-age disability. Journal of Gerontology: Social Sciences, 56B, pp.
to Working Life. Final Comparative Report, DEELA/ED (99) 11.
Paris: Directorate for Education, Employment, Labour and Social Schooler, C., M. S. Mulatu, and G. Oates (1999) The continuing effects Affairs, Education Committee, Organisation for Economic Co- of substantively complex work in the intellectual functioning of older workers. Psychology and Aging, 14(3), pp. 483–506.
OECD (2006) Live longer, work longer: A synthesis report. Available Schwartzman, A. E., D. Gold, D. Andres, T. Y. Arbuckle, and J.
online at www.oecd.org/document/42/0,3343,en_2649_34747 Chaikelson (1987) Stability of intelligence. A 40 year follow-up.
Canadian Journal of Psychology, 41, pp. 244–256.
Osler, W. (1906) The Fixed Period. (3rd chapter in Aequanimitas). New Scott, F., M. Berger, and J. Garen (1995) Do health insurance and pension costs reduce the job opportunities of older workers? Oster, S. and D. S. Hamermesh (1998) Aging and productivity among econ- Industrial and Labour Relations Review, 48, pp. 775–791.
omists. The Review of Economics and Statistics, 80(1), pp. 154–156.
Simonton, D. K. (1988) Age and outstanding achievement: What do we Park, D. C., R. E. Nisbett, and T. Hedden (1999) Culture, cognition, know after a century of research? Psychological Bulletin, 104, pp.
and aging. Journal of Gerontology, 54B, pp. 75–84.
Parkin, T. (2003) Old Age in the Roman world. A cultural and social Skirbekk, V. (2008) Productivity potential during the life cycle: The history. Baltimore, MD: Johns Hopkins University Press.
changing importance of age-specific abilities. Population and Phelps, R. and J. Shanteau (1978) Livestock judges: How much Development Review, 34, pp. 191–207.
information can an expert use? Organizational Behavior and Human Smith, A. D. (1996) Memory. In: J. E. Birren and K. W. Schaie (eds.) Performance, 21, pp. 209–219.
Handbook of the Psychology of Aging. New York: Academic Press, Prskawetz, A., B. Mahlberg, and V. Skirbekk (Forthcoming 2007) Firm productivity, workforce age and educational structure in Austrian indus- Solem, P. (2006) Age and context in subjective work ability changes.
tries. In A. Mason and R. Clark (eds.) Intergenerational Transfers and the Macroeconomy. Cheltenham: Edward Elgar.
Solfrizzi, V., F. Panza, and F. Capurso (2003) The role of diet in cogni- Prskawetz, A., B. Mahlberg, V. Skirbekk, I. Freund, M. Winkler-Dworak, tive decline. Journal of Neural Transmission, 110(1), pp. 95–110.
T. Lindh, B. Malmberg, A.-C. Jans, O. S. Nordström, and F. Andersson Spitz-Oener, A. (2006) Technical change, job tasks, and rising (2005) The impact of population ageing on innovation and productivity educational demands: Looking outside the wage structure. Journal of growth in Europe. Brussels: European Commission, European Labor Economics, 24(2), pp. 235–270.
Observatory on Demography and the Social Situation (SSO), Studies on Tang, J. and C. MacLeod (2006) Labour force ageing and productivity policy implications of demographic changes in the enlarged EU.
performance in Canada. Canadian Journal of Economics, 39(2), pp.
Raskin, E. (1936) Comparison of scientific and literary ability: A biographical study of eminent scientists and letters of the nineteenth Tuddenham, R. D. (1948) Soldier intelligence in World Wars I and II.
century. Journal of Abnormal and Social Psychology, 31, pp. 20–35.
American Psychologist, 3, pp. 54–56.
Remery, C. K., J. Henkens, J. Schippers, and P. Ekamper (2003) Managing U.S. Department of Labor (1957) Comparative job performance by age.
an aging workforce and a tight labor market: Views held by Dutch Large plants in the men’s footwear and household furniture industries.
employers. Population Research and Policy Review, 22, pp. 21–40.
Bulletin No. 1223. Washington, D.C.: U.S. Department of Labor.
Riphahn, R. T. and P. Trübswetter (2006) Population Aging and van Ours, J., P. Ilmakunnas, V. Skirbekk, and M. Weiss (2007) Aging and Productivity. Paper presented at the IX Conference Fondazione Romeu Gordo, L. (2005) Compression of Morbidity and the Labor Supply of Older People. Working Paper. Berlin: Max Planck Verhaegen, P. and T. A. Salthouse (1997) Meta-analyses of age- cognition relations in adulthood. Estimates of linear and nonlinear Rubin, D. K. and J. M. Perloff (1993) Who works for piece rates and why.
age effects and structural models. Psychological Bulletin, 122(3), pp.
American Journal of Agricultural Economics, 627, pp. 1036–1043.
Rybash, J. M., W. Hoyer, and P. A. Roodin (1986) Adult Cognition Waldman, D. A. and B. J. Avolio (1986) A meta-analysis of age and Ageing. New York: Pergamon Press.
differences in job performance. Journal of Applied Psychology, 71, Salthouse, T. (1984) Effects of age and skills in typing. Journal of Experimental Psychology, 113, pp. 345–371.
Warr, P. (1994) Age and employment. In: H. Triandis, M. Dunnette, Salthouse, T. and T. J. Maurer (1996) Aging, job performance and and L. Hough (eds.) Handbook of Industrial and Organizational career development. In: J. E. Birren and K. W. Schaie (eds.) Psychology. Thousand Oaks, California: Consulting Psychologist Handbook of the Psychology of Aging. New York: Academic Press, Zwick, T. (2002) Continuous Training and Firm Productivity in Schaie, K. W. and S. L. Willis (1986) Can decline in intellectual func- Germany. ZEW Discussion Paper No. 02–50. Mannheim, Germany: tioning be reversed? Developmental Psychology, 22(2), pp. 223–232.
Centre for European Economic Research.
CorrespondenceWorld Population Program, International Institute for Applied Systems Analysis (IIASA), Schlossplatz 1, A-2361 Laxenburg, AustriaEmail: [email protected] AGEING HORIZONS Issue No 8
OXFORD INSTITUTE OF AGEING
Annex: Overview of employer-employee data sets
Sample size
Control variables
Productivity
Age-productivity
measurement
Productivity peaks between Manufacturing and capital and input factorsR&D spending Gender, race, occupation, Firm’s output or value AGEING HORIZONS Issue No 8
OXFORD INSTITUTE OF AGEING

Source: http://user.iiasa.ac.at/~skirbekk/ageing_horizons_8_skirbekk_ll.pdf