Jump to ContentJump to Main Navigation
Show Summary Details
More options …

The B.E. Journal of Economic Analysis & Policy

Editor-in-Chief: Jürges, Hendrik / Ludwig, Sandra

Ed. by Auriol , Emmanuelle / Brunner, Johann / Fleck, Robert / Mendola, Mariapia / Requate, Till / Schirle, Tammy / de Vries, Frans / Zulehner, Christine

4 Issues per year

IMPACT FACTOR 2016: 0.252
5-year IMPACT FACTOR: 0.755

CiteScore 2016: 0.48

SCImago Journal Rank (SJR) 2016: 0.330
Source Normalized Impact per Paper (SNIP) 2016: 0.526

See all formats and pricing
More options …
Volume 13, Issue 2


Volume 6 (2006)

Volume 4 (2004)

Volume 2 (2002)

Volume 1 (2001)

Age, Human Capital, and the Quality of Work: New Evidence from Old Masters

Thomas Bayer / John Page / Yaron Raviv
  • Corresponding author
  • Robert Day School of Economics and Finance, Claremont McKenna College, 500 E. Ninth St. Claremont, CA 91711, USA
  • Email
  • Other articles by this author:
  • De Gruyter OnlineGoogle Scholar
/ Joshua Rosett
Published Online: 2013-07-03 | DOI: https://doi.org/10.1515/bejeap-2012-0030


The links between individual ability, human capital investment, and quality of output are generally hard to examine because in most situations output results from multiple inputs and often through complex contracting processes. We overcome these problems by examining life-cycle artistic output quality as reflected in art auction prices. First, we observe an inverted U-shaped age-quality of work profile similar to the conventional age–wage profile. Second, we find that the degree of concavity increases for those with higher native ability. Third, we find that working for a patron rather than selling directly to the market is associated with a flatter age profile. Fourth, we find evidence that formal education increases the concavity of the age-quality of work profile. These results are consistent with the theory and demonstrate that artists respond to incentives to invest in human capital.

Keywords: art auctions; quality of work; age

JEL Classification: J24; N33; D44


  • Abraham, Katharine G., and Henry S. Farber. 1987. “Job Duration, Seniority, and Earnings.” American Economic Review 77 (3):278–97.Google Scholar

  • Accominotti, Fabien. 2009. “Creativity from Interaction: Artistic Movements and the Creativity Careers of Modern Painters.” Poetics 37:267–94.Web of ScienceCrossrefGoogle Scholar

  • Acemoglu, Daron, and Jorn-Steffen Pischke. 1999. “Beyond Becker: Training in Imperfect Labour Markets.” Economic Journal 109:112–42.Google Scholar

  • Autor, David H. 2001. “Why Do Temporary Help Firms Provide Free General Skills Training?” Quarterly Journal of Economics 116 (4):1409–48.Google Scholar

  • Bayer, Thomas M., and John R. Page. 2011. The Development of the Art Market England: Money as Muse, 1730–1900. London: Pickering & Chatto Publishers.Google Scholar

  • Beggs, A., and Kathryn Graddy. 1997. “Declining Values and the Afternoon Effect: Evidence from Art Auctions.” Rand Journal of Economics 28:544–65.CrossrefGoogle Scholar

  • Bredius, Abraham. 1891. De kunsthandel te Amsterdam in de XVIIe eeuw. Amsterdam: Jaarboekje.Google Scholar

  • Chong, Alan. 1987. “The Market for Landscape Painting in Seventeenth-Century Holland.” In Masters of the Seventeenth Century Dutch Landscape Painting, edited by Peter Sutton, 104–20. Boston, MA: Boston Museum of Fine Art.Google Scholar

  • Diamond, Arthur M., Jr. 1986. “The Life-Cycle Research Productivity of Mathematicians and Scientists.” Journal of Gerontology 41 (4):520–5.CrossrefGoogle Scholar

  • Dohmen, Thomas J. 2004. “Performance, Seniority, and Wages: Formal Salary Systems and Individual Earnings Profiles.” Labour Economics 11:741–63.CrossrefGoogle Scholar

  • Fair, Ray. 1994. “How Fast Do Old Men Slow Down?” The Review of Economics and Statistics 76:103–18.Google Scholar

  • Galenson, David W., and Bruce A. Weinberg. 2000. “Age and the Quality of Work: The Case of Modern American Painters.” Journal of Political Economy 108 (4):761–77.Google Scholar

  • Galenson, David W., and Bruce A. Weinberg. 2001. “Creating Modern Art: The Changing Careers of Painters in France from Impressionism to Cubism.” American Economic Review 91 (4):1063–71.Google Scholar

  • Ginsburgh, Victor, and Sheila Weyers. 2006. “Creativity and Life Cycles of Artists.” Journal of Cultural Economics 30:91–107.CrossrefGoogle Scholar

  • Graves, Algernon. 1970. Art Sales from Early in the Eighteenth Century to Early in the Twentieth Century. New York: Burt Franklin, 3 volumes, originally published in 1918; and George Redford, Art Sales: A History of Sales of Pictures & Other Works of Art, London: 1888.Google Scholar

  • Hellmanzik, Christiane. 2009. “Artistic Styles: Revisiting the Analysis of Modern Artists’ Careers.” Journal of Cultural Economics 33:201–32.Web of ScienceGoogle Scholar

  • Hellmanzik, Christiane. 2010. “Location Matters: Estimating Cluster Premiums for Prominent Modern Artists.” European Economic Review 54:199–218.Web of ScienceGoogle Scholar

  • Hutchens, Robert M. 1989. “Seniority, Wages, and Productivity: A Turbulent Decade.” Journal of Economics Perspectives 3:49–64.Google Scholar

  • Lazear, Edward P. 1976. “Age, Experience, and Wage Growth.” American Economic Review 66 (4):548–58.Google Scholar

  • Lazear, Edward P. 1979. “Why Is There Mandatory Retirement?” Journal of Political Economy 87 (6):1261–84.Google Scholar

  • Levin, Sharon, and Paula Stephan. 1991. “Research Productivity Over the Life Cycle: Evidence for Academic Scientists.” American Economic Review 81:114–32.Google Scholar

  • Montias, John Michael. 1983. Artists and Artisans in Delft: A Socio-Economic Study of the Seventeenth Century. New York: Princeton University Press.Google Scholar

  • Montias, John Michael. 1985. “Flemish and Dutch Trade in Works of Art in the 16th and 17th Centuries.” Institution for Social and Policy Studies, Working Paper No. 1022, Yale University.Google Scholar

  • Montias, John Michael. 1987. “Cost and Value in Seventeenth-Century Dutch Art.” Art History 10:455–66.CrossrefGoogle Scholar

  • Montias, John Michael. 1988. “Art Dealers in the Seventeenth-Century Netherlands.” Simiolus 18 (4):244–56.CrossrefGoogle Scholar

  • Montias, John Michael. 1990. “Socio-Economic Aspects of Netherlandish Art from the Fifteenth to The Seventeenth Century: A Survey.” The Art Bulletin LXXII (3):358–73.Google Scholar

  • Oster, Sharon M., and Daniel S. Hamermesh. 1998. “Aging and Productivity Among Economists.” Review of Economics and Statistics 80 (1):154–6.Google Scholar

  • Raviv, Yaron. 2006. “New Evidence on Price Anomalies in Sequential Auctions: Used Cars in New Jersey.” Journal of Business and Economic Statistics 24:301–12.CrossrefGoogle Scholar

About the article

Published Online: 2013-07-03

For a literature survey see Hutchens (1989).

For example, see Lazear (1976, 1979).

We are greatly indebted to Christie’s head archivist, Jeremy Rex-Parkes, and his staff. Their warm welcome, research-supportive attitude and good English humor were exceptionally helpful as was, of course, the unimpeded access to the archive’s collection of catalogues going back to the 1760s.

As explained in Bayer and Page (2011), Graves’ records are generally used for provenance research, as they list records under the alphabetically ordered artists’ names, the date of sale, the auction house conducting the sale, the name of the seller, lot number, painting title, date, medium, size, buyer’s name, and finally the auction price. Other commonly used sources include Bredius, Montias, and Chong, all well-known to art historians. Abraham Bredius compiled the first comprehensive inventory of Dutch collections in the seventeenth century (Bredius 1891). John M. Montias has published a number of articles and books on the art market in seventeenth century Holland (Montias 1983, 1985, 1987, 1988, 1990). Alan Chong, formerly director at the Gardener Museum and now director of the Asian Civilizations Museum in Singapore, did an innovative study on the market for landscape paintings in seventeenth century Holland (Chong 1987). Our transaction records contain far more information than Bredius’ inventories or the inventory analyses undertaken by J. M. Montias or Alan Chong.

In the process of verifying and extending the Graves data, we constructed a separate database primarily from the Christie’s records. In the summer of 2000, this database was acquired by the French art market indexing company, artprice.com, under the title Bayer/Page Data Base. In the fall of 2000, artprice.com announced the availability to its subscribers of econometric analyses of the present-day art market based on our analytical models. For a complete description of the construction and testing of the data used here, see Bayer and Page (2011).

Use of logs is standard in existing literature on this subject. For example, see Galenson and Weinberg (2000, 2001).

We report real prices in the body of the paper based on the price level in 1700. To provide an idea of current value, we use Bank of England price index information to convert our prices to current values. The adjustment factor from 1700 to 2008 is around 130. Based on this, for example, the mean selling price of £303.81 is worth £39,485 in 2008 prices. The most expensive work (Madame Pompadou by Francois Boucher) would cost £949,098 in 2008 prices. This pricing estimate is based solely on the Bank of England price index and it might be different than an “art inflation index.” We do not have information on the selling price of this particular painting in recent years. It was donated by Sir Richard Wallace’s widow to the Wallace Collection in London after his death in 1897. It has remained there since then.

In the full sample, Sir Joshua Reynolds appears most frequently (2,423 observations).

Qualitatively similar results were obtained when we exclude this variable from our analysis.

There may be other factors that could affect the selling price in an auction such as the item order of the sale. This information is beyond the scope of our dataset, however. See Raviv (2006) for further discussion on the effect of the item order of sale on the selling price. As robustness checks, we also included in unreported regressions variables including the number of times the seller appears in our database, the number of times the buyer appears in our database and the square of the age of the work. We also ran a random effects version of the reported results, which are qualitatively similar with respect to the variables of interest. We also ran a weighted version of our regressions using the Artist Observations variable as weights, and obtained qualitatively similar results.

It is not clear whether increasing age of the painting during the time period we examine should increase or decrease prices. Value as an antiquity may rise over time, but in the absence of modern restoration techniques value may fall due to deterioration. Beggs and Graddy (1997), for example, find that there is a negative impact of painting age on price. However, this is more pronounced for contemporary art sample and less so for impressionist and modern art. We believe the explanation for the difference in their results could be that the former sample is more likely to include living artists, where the supply of art by the given artist is increasing over time. For the earlier sample, the supply is more likely fixed. Our sample consists entirely of artists with (now) fixed supply of artworks.

The average year of execution for Galenson and Weinberg (2001) earlier group is 1893, which is still considerably later than our average execution year of 1847.

An alternative explanation might be that those with higher talent have a higher variance in quality throughout their lives. This could even be consistent with artists that have a formal education having a steeper quality profile than artists without a formal education. We thank an anonymous referee for suggesting this alternative explanation for our findings.

This finding is somewhat similar to the empirical evidence of different age-wage profiles between white collar and blue collar. Roughly speaking, the coefficient on the age variable is positive and statistically significantly higher for the white collar workers than for blue collar workers. The coefficient on the age squared variable is negative and statistically significant for both white collar and blue collar workers. The differences between the two groups with respect to the age-squared variable is mixed (it could be greater, similar or smaller). See, for example, Abraham and Farber (1987) who demonstrate it using data from the Panel Study of Income Dynamics, and Dohmen (2004) who demonstrates it using data from personnel files of a Dutch aircraft manufacture.

Co-author Thomas Bayer coded our measure of talent. Bayer has extensive experience as an art dealer in New Orleans, and contributed his judgments regarding talent of artists. To the extent possible, he used Royal Academy membership as a proxy for talent. This seems reasonable given that admittance to the Academy required scrutiny by experts of the day, and is accepted by art dealers today as evidence of quality. There may have been institutional biases due to restrictions on Academy membership, such as race or religious restriction. Hence, first we believe his judgment helps overcome biases on Academy membership. Second, Royal Academy membership is not applicable to our entire sample as the Royal Academy was founded in 1768 and would apply to Old Masters English but not to Old Master Continental. (We make the distinction between Old Master English (OME) and Old Masters Continental (OMC); Contemporary British (CB) and Contemporary Continental (CC).) The OME category lists painters of English origin or resided in London. Among them are several who were members of the Royal Academy and thus they fit into “talented or native ability” group. No painter listed in the OMC category has Royal Academy affiliation. On the Continent, from where Continental Masters originated, there were no such vetting organizations at the time these painters were active. Hence we rely on Bayer’s judgments regarding those not eligible for the Academy due either to residency or time period. We note that his judgments were applied before observing any empirical results to minimize the chance of inadvertent bias.

We thank the editor for suggesting this alternative procedure. In principle, a complete set of interactions of the classification variables with all regressors should be included in order to replicate the Table 3 results for the regressions run separately by classification group. However, this is not feasible due to the large number of regressors (including indicator variables and decade dummies) in our specifications. Hence it is likely that the lack of significance on the main effects for age is partly due to this misspecification.

Other evidence in the economic literature suggest that high ability workers tend to invest and train more than low ability workers. For example, Autor (2001) describe a model in which firms offer general training to induce self-selection and perform screening of the worker ability. In the model, training attracts high ability workers. He found empirical evidence for this using the Bureau of Labor Statistics 1994 Occupational Compensation Survey of Temporary Help Supply Services. Acemoglu and Pischke (1999) provide a broad survey regarding training.

Citation Information: The B.E. Journal of Economic Analysis & Policy, Volume 13, Issue 2, Pages 687–708, ISSN (Online) 1935-1682, ISSN (Print) 2194-6108, DOI: https://doi.org/10.1515/bejeap-2012-0030.

Export Citation

©2013 by Walter de Gruyter Berlin / Boston. Copyright Clearance Center

Citing Articles

Here you can find all Crossref-listed publications in which this article is cited. If you would like to receive automatic email messages as soon as this article is cited in other publications, simply activate the “Citation Alert” on the top of this page.

Doron Kliger, Yaron Raviv, Joshua Rosett, Thomas Bayer, and John Page
Journal of Behavioral and Experimental Economics, 2015, Volume 59, Page 74

Comments (0)

Please log in or register to comment.
Log in