VoxEU Column Frontiers of economic research Labour Markets

Titian, Veronese, Caravaggio and their rivals: Evidence of competition in the market for altarpieces of the 17th century

Looking at the contracts for large oil paintings in Italy (1550-1750), this column finds evidence of strong competition between painters. Contracts were structured to address moral hazard problems, and prices closely reflected demand and supply conditions in an integrated market.

Art is often perceived, and sometimes defined, as handmade work that is valuable independently of its objective features and is the fruit of pure talent and inspiration independent from monetary and contractual incentives. This emphasis on the invaluable is even more pronounced for religious paintings by old masters as Titian, Veronese, or Caravaggio. At the same time, the pricing of a unique art object is often perceived as highly subjective and largely dependent on the tastes, wealth and prestige of buyers, with little regard for factors affecting demand and supply. This is perception is especially strong when considering the Renaissance and Baroque periods, in which honour and prestige were claimed to be the drivers of social and economic activities (of patrons and artists) more than the profit-seeking behaviour of the “homo economicus”.

Exhibit 1. Paolo Veronese, Wedding at Cana, Paris, Louvre Museum ©

Even a look at the prices of famous old paintings may reinforce the impression that little can be rationalised in this market. The Wedding at Cana by Veronese (Exhibit 1) was commissioned in 1563 for a wall of the nave of the Venetian church of S. Giorgio Maggiore for 324 silver ducats. It is, and was at the time, considered a masterpiece for its imposing composition, the splendid contemporary costumes, and the luminous colours, probably not because of its size (about 70 square meters) or for the number of human figures depicted (more than fifty major ones), which were largely agreed with the commissioners.

Today we find it at the Louvre Museum, right in front of another painting whose size is much smaller, but whose value is hardly so: the portrait of a single human figure, the Mona Lisa by Leonardo Da Vinci. Less known than the original painting by Veronese are its copies. One of these, by the minor painter Zanchi for a private collection, was commissioned for 1,000 ducats in 1671 – more than the double that of the original, even taking inflation into account.

Another of the three great masters of the 16th century in Venice, Jacopo Tintoretto, had a long and brilliant career, but he never managed to be paid as much as the rival Titian for his works. In 1583 he reached his highest fee (400 ducats) for a Nativity commissioned for the Escorial building by the king of Spain, Philip II. Apparently, Tintoretto did not even put much effort in this work, since his less talented son Domenico painted most of it. In 1625, an art dealer who was contracting an altarpiece by Cerano in Milan told the patron that the painter would have probably accepted about 250 ducats, but also that if Cerano were to go to Rome he would be paid double because, he added, Rome is "where you go to get rich". Yet at the beginning of the century, the fees for the revolutionary paintings by Caravaggio were extremely low. Meanwhile, the more fashionable Guido Reni was starting a successful career that would have made him the best-paid painter of the century in Italy. One of his last altarpieces, the Adoration of the Shepherds (1640) for the Certosa of San Martino in Naples, was paid the record price of more than 3,000 ducats. Each figure depicted cost more than an average altarpiece by ordinary painters, including Caravaggio, whose fame came only centuries later.

The logic of paintings’ prices

In spite of all these confusing anecdotes, a look at the contracts on commissions for oil paintings of historical subjects (religious and mythological) between 1550 and 1750 (from the late Mannerism to the entire Baroque age and the Rococò period), can provide a different view. Building on recently collected data by the historical research of Spear and Sohm (2010), our research suggests that economic conditions and profit-maximising behaviour were in fact crucial determinants of the contracts between these painters and their patrons, and that strong competition between painters affected equilibrium prices, as we could expect in a fully fledged market economy. We know this because most commissions were formalised in detailed contracts signed in front of notaries with validity throughout Italy, defining the price and the mutual responsibilities of the artist with his workshop and of the patron. A lot of these contracts (or related documentary evidence) have survived, allowing Laura Pagani and I to analyse this market with econometric techniques (see Etro and Pagani 2010).1

We found one of the first markets in modern history for which the basic laws of economics and rational behaviour apply. Indeed, we find they explain the prices of these paintings with remarkable precision. After taking into account obvious price differences due to the different quality of painters as perceived at the time (whose proxy was their average annual income), we find a number of relationships between the price of paintings and quantifiable features often linked to demand and supply conditions. For instance, we find a positive and concave relationship between prices and size of the paintings, which reflects economies of scale in their production. Ceteris paribus, each additional square metre increases the price of the average painting by 9%. Other relevant explanatory variables include the placement of religious paintings in the churches (demand was more rigid for altarpieces, commanding higher prices, and more elastic for paintings on the nave, which could be substituted with different decorations), or the institutional nature of the commissioner (the Ducal Palace in Venice or St Peters in Rome wanted and obtained more quality by paying artists a little more than the ordinary market – a sort of efficiency wage mechanism).

Solving art’s moral hazard problem

More surprisingly, we find evidence of simple solutions to the moral hazard problem emerging between patrons and artists. Large commissions for oil paintings of historical subject required months or years of work and generated conflicts of interest for the simple reason that quality required time and effort – but this could not be defined in a contract.2

We support the idea that patrons and artists adopted a typical solution to the moral hazard problem pointed out in the standard literature on principal-agent contracts (the informativeness principle of Holmstrom 1979). Prices were made conditional on measurable features of the paintings that were positively correlated with quality. In the case of historical paintings, this was possible through the number of human figures (as the more than fifty present in the Veronese’s paintings), which was not equivalent to the absolute quality of a painting, but was correlated with it for at least three main reasons.

  • First, the subjects were biblical or mythological stories of man, women, saints, angels or mythological gods, where imagination and storytelling had a crucial function. Therefore, it was safe to conclude that the variety and complexity of the composition, summarised by the number of players, had a positive correlation with quality.
  • Second, at the time there was a precise ranking in the aesthetic evaluation of subjects: historical compositions were best, followed by portraits, landscapes, and finally still lifes. A higher number of human figures reduced the space available for other decorative subjects, automatically enhancing total quality.
  • Third, painters were often focused their effort on human figures, delegating less relevant parts to their assistants. Accordingly, a higher number of figures was a proxy for a wider direct intervention of the painters.

The data show that, ceteris paribus, one more figure in the painting resulted in an increase of the price of 3% in Venice and up to 16 % in the rest of Italy (notoriously, colour was more important in Venice and figure drawings in Florence and Rome).

The wider composition: Endogenous entry of painters in local markets

Moving to macroeconomic aspects, we evaluate the impact of local demand shocks and aggregate demand shocks. Differences in local demand could be detected when looking at different destinations: demand was higher in big and rich cities as Venice compared to small provincial towns in the countryside. However, the mobility of painters was high. Therefore we should expect that price differentials between high-demand and low-demand towns should be arbitraged away. Indeed, we find that prices in the countryside were much lower than in Venice, but, after controlling for paintings' and painters' features, all these differences disappear.

Paintings exported abroad earned more, simply because they were of different quality and painted by better artists than the others. Identical results emerge when we look at Rome, which was the leading international artistic centre during the Baroque age. Its prices were higher, but only because the best painters moved there. Taking into account all the differences, there was no way for a painter to earn more by changing the destination of his works. This suggests that entry in the local markets was endogenous and the opportunities for extra profits were eliminated through the mobility of painters.

While market forces appear to have been at work to induce price equalisation in a largely integrated market, aggregate demand shocks exerted direct effects on prices. As an example, we study the impact of the plague, which hit Venice and all the surrounding regions in 1630, and we verify that its impact was to reduce prices in a significant way. The same happened in the other art centres of Florence, Rome, and Naples, where the plague arrived in 1656.

Our study shows that, far from being a world beyond the vagaries of the market, the market for oil paintings in 1550–1750 in fact offers one of the earliest concrete examples of the laws of economics in action.


Etro, F and L Pagani (2010), “The Market for Paintings in Baroque Venice”, WP 191, University of Milan, Bicocca, Department of Economics.

Holmstrom, B (1979), “Moral Hazard and Observability”, Bell Journal of Economics, 10(1):74-91

O'Malley, M (2005), The Business of Art. Contracts and the Commissioning Process in Renaissance Italy, Yale University Press

Spear, R and P Sohm (2010), Painting for Profit: The Economic Lives of Seventeenth-Century Italian Painters, Yale University Press

1 Prices were in silver coins in each town and their value was approximately equal. In the empirical analysis all the prices were normalized for the cost of living, expressed through the price of wheat.
2 To induce effort, multiple contractual solutions were adopted since Renaissance (see O’Malley, 2006). Many contracts required preliminary drawings to be evaluated and possibly approved by the commissioners. Rejection of the painting in case of low quality was a credible threat for the artists. However, both these practices could only insure a minimum level of effort. Finally, contracts occasionally left space for bonuses for quality, but the judgement was often by external painters, inducing conflicts of interests. The last practice may be seen as a sort of incentive contract, but its effectiveness appears to have been quite limited.

10,499 Reads