The Damage Inflicted by French Rent Controls between 1914 and 1948

In an essay in Metropolitics, Loïc Bonneval questioned the consensus among economists that strict rent controls have perverse effects. Bonneval based his analysis on a study of apartment buildings in Lyon between the two world wars. Robert Ellickson and David Le Bris challenge Bonneval’s interpretation, and marshal evidence of the damage that rent controls inflicted in Lyon and France between 1914 and 1948.

In an article originally published in 2011 in Métropolitiques (in French) and recently translated into English in Metropolitics, sociologist Loïc Bonneval asserted that France’s severe rent controls between 1914 and 1948 did not impair the profitability of investments in rental housing. He did not contest the severe housing shortage that the French endured between the two wars, nor the plunge in housing construction. But, in claiming that the profitability of real-estate investments was not so bad, Bonneval implicitly opened the way for new experiments in rent control. While we applaud the quality of Bonneval’s historical research and his willingness to challenge conventional thinking, we assert that his economic analysis is unsound. More specifically, the analysis rests on an issue of minor importance: the profits available to someone who purchases a building when rents already are controlled. Far more important are the profits available to an existing landlord, or to a would-be builder of rental housing. Strict rent controls draw political support because they confer various benefits on sitting tenants, such as greater security of continued occupation. The innumerable costs of rent controls typically more than offset these advantages.

The harmful effects of strict rent controls: the economic consensus

In a 40-question poll of US economists, the proposition that “a ceiling on rents reduces the quantity and quality of housing available” attracted more support than any other. Some 76.3% of respondents “generally agreed” with the statement, while an additional 16.6% “agreed with reservations” (Alston et al. 1992, p. 204). In a similar poll of French economists, when asked the question of whether “rent controls reinforce a shortage of housing,” 21% “absolutely agreed” and 45% “somewhat agreed,” while only 19% “somewhat disagreed” and 7% “absolutely disagreed” (Wasmer and Mayer 2009, q. 57). Because economists disagree on most subjects, this widespread agreement is surprising. It is precisely this consensus that Loïc Bonneval challenges.

The influence of Paul Samuelson helps explain why US economists are particularly hostile to rent controls. Throughout his career, Samuelson, one of the most eminent economists of the 20th century and a political centrist, often invoked interwar rent controls in France as a prime example of wrongheaded lawmaking. [1] Perhaps on account of Samuelson, references to this episode in French history have long been a staple in the teaching of microeconomics in the United States. Ironically, this history seems to be less well known in France, even in economics departments.

In 1948, France finally relaxed the strict regulation of rents that had prevailed since 1914. [2] By 1948, French residential tenants were paying on average 2% of their income on rent, compared to between 15% and 20% in 1914. [3] In a 1949 speech, Eugène Claudius-Petit, the minister for reconstruction and urban development, asserted that, in the previous year, French households had spent seven times more on tobacco than on rent. As classic economic theory predicts, landlords responded to rent controls by reducing their outlays for maintenance, potential investors in housing became less likely to provide financing, and tenants found their mobility impaired.

Profitability, but for whom?

Bonneval concentrates his attention on the effects of rent controls on the profits available to a purchaser of a building whose rents already are controlled. [4] This focus misses the mark. A buyer, if aware that rents are controlled, can assure profitability by buying at a lower price. Bonneval recognizes that, in fact, this is what happened in Lyon. [5]

Much more important, in terms of economic consequences, are the effects of rent controls on the profits obtained by existing owners of buildings, and, in particular, by potential builders of new rental housing.

From 1914 on, rent controls inflicted catastrophic losses on owners of apartment buildings. Bonneval and Robert’s book (2013), which analyzes the records of a firm that managed dozens of these structures in Lyon between 1870 and 1968, confirms this fact. It shows that, between 1913 and 1948, the real (inflation-adjusted) value of apartment buildings in Paris and Lyon fell by 90%. [6] Thomas Piketty, in an independent analysis, confirms this 90% falloff. [7]

Further proof of the magnitude of these losses is furnished by the history of La Fourmi Immobilier, a firm that owned the same 15 Paris apartment buildings for most of the 20th century. The firm owed no debt and distributed in dividends almost all the net rents it received. After rent controls were enacted, these dividends fell drastically in real terms. Compared to 1914, they had fallen by 50% in 1939, and by 95% in 1948. As a consequence, at the Paris Bourse, the value of stock in La Fourmi dropped, in 1939, to 8% of its value before World War I, and, in 1948, to 6% of its prewar value (Simonnet et al. 1998, p. 128).

The dramatic fall in the value of apartment buildings between the two wars validates the conventional economic theory of the effects of price controls. The proprietor of a going concern, such as an apartment building (or a share of its stock), expects to receive not only a flow of revenue from the enterprise (or personally share in its use), but also the value of the enterprise when it is eventually sold. The market value of a going concern is the sum of these values, discounted to the present. Rent controls reduce the present value of an existing apartment building by lowering both the anticipated rental income and the anticipated profits at time of resale. Conversely, an unexpected loosening of rent controls—for example, the French reforms in 1948—is likely to increase the value of existing structures.

By 1938, the real value of the rents that La Fourmi received was 42% of what it had been before 1914, and, by 1948, less than 5%. Bonneval and Robert show collapses of a similar magnitude in Lyon (2013, p. 88). This was hardly a profitable time to be a landlord in France.

Landlords responded to these losses in revenue by slashing expenditure on maintenance. Rent-controlled dwellings tend to be poorly maintained. [8] Controls reduce not only the financial resources of landlords, but also the incentive to fix their buildings. Tenants benefiting from a reduced rent are likely to remain in residence whatever the condition of the premises. [9] Tellingly, Bonneval and Robert agree that French apartment buildings suffered from a lack of maintenance during the interwar period. [10]

Putting the brakes on new construction

More importantly for the well-being of the French population as a whole, the strictness of interwar rent controls discouraged the construction of new rental housing. An investor considering building a new apartment building had to anticipate receiving both less rental income and a lower price at the time of resale. During the interwar period, the costs of construction rose on account of inflation in the price of both construction materials and workers’ wages, neither of which were subject to price controls. Measured in constant francs, the construction cost of a standard apartment building increased by around 40% between 1914 and 1938 (Bonneval and Robert 2013, p. 224), while the real value of rental income fell by about 60% (p. 88). Under these conditions, most investors understandably refused to finance the construction of new apartment buildings. [11] A buyer of an existing apartment building could avoid the impact of rent controls by bidding less, exactly what Bonneval observed in Lyon. But a builder of a new apartment building had could not avoid rising construction costs.

Before the creation of INSEE (France’s national statistics office) in 1946, reliable data on the construction of multifamily structures were lacking. This makes it difficult to estimate how much—and why—apartment construction declined in France. [12] Nevertheless, two estimates of the magnitude of the interwar production of dwellings support the idea that rent controls dampened construction activity during an era when a large majority of households were renters. An INSEE report indicates that 1.8 million dwellings of all types were built in France between 1919 and 1939. [13] And, for the same period, another official document reports a lower figure: 1.5 million dwelling units. [14] By comparison, during the same time period about 4 million dwellings were constructed in Germany and 3.7 million in Great Britain. [15] A majority of analysts have attributed a large part of France’s anemic level of interwar housing production to the nation’s rent controls. [16]

The impairment of tenant mobility

Numerous studies, in France and elsewhere, confirm that rent controls lessen the likelihood that a tenant will move to another address. [17] To avoid losing the advantage of a below-market rent, a worker might refuse a better job in a more remote location, or a grandparent might decline to move closer to children and grandchildren. Bonneval and Robert themselves provide data showing that the annual turnover of tenants in Lyon fell from around 12% per year before 1914 to 5% per year in the 1920s and 1930s (2013, p. 130).

More generally, strict rent controls reduce vacancy rates and interfere with the interplay of forces of supply and demand. “No Vacancies” was the title that Bertrand de Jouvenel (1948) chose for his polemic against France’s rent controls. Bonneval and Robert confirm that the apartment vacancy rate in Lyon was higher before World War I than afterward (pp. 125–126). Artificially low rents can induce a tenant to remain in a house or apartment that they otherwise would have abandoned (Glaeser and Luttmer 2003). Thus, when rent controls are strict, an apartment with three bedrooms may end up being occupied by a single person.

The costs of administering rent controls

The implementation of rent controls consumes time and money. These costs are borne by the taxpayer, who finances the courts and public agencies that manage rent controls, and by landlords and tenants themselves. Once rent control has been instituted, it is politically very difficult to repeal, even when its negative effects are obvious. In France, only the severe disruptions of World War II made the unblocking of rents possible. When reforming the system in 1948, the French legislature revealed itself to be incapable of completely sweeping away the past. In its effort to phase out rent controls, the law of 1948 created an incredibly complex administrative mechanism for determining an allegedly “scientific rent.” This Kafkaesque system endeavored to valuate each square meter of a rental dwelling, taking into account multiple variables such as sunlight, proximity to stores, and access to utility services. One can readily imagine the conflicts between landlords and tenants that these procedures created. For some dwellings, the calculation of a “scientific rent” continues to this day.

The empirical work that Bonneval and Robert carried out is remarkable. Their study of rental buildings in Lyon invaluably contributes to a better understanding of the economics and management of real estate. Bonneval takes meticulous care when describing the positions of the many researchers with whom he disagrees. And the book is a treasure trove of citations that point to valuable sources on an important and relatively neglected episode in the economic and political history of France.

On the other hand, though, the interpretation of this history is suspect. It relies on a peripheral issue to challenge the economic consensus that strict rent controls have deleterious effects. The history of French housing between 1914 and 1948 in fact provides a textbook case against rent controls, as Paul Samuelson recognized.

The economic shocks of the 20th century have prompted state authorities to intervene profoundly in economic activity, including, notably, in landlord–tenant relations (Voldman 2016). Some rent controls, mild in comparison to those in force in France during the interwar period, continue to exist in New York City and Germany. [18]

At the outset of the 21st century, political conditions have again prompted proposals to strengthen rent regulation. Policymakers should be skeptical. They should keep in mind the damage done in the past by policies that, at the time, were widely regarded as well intentioned.


  • Alston, R., Kearl, J. and Vaughan, M. 1992. “Is There a Consensus Among Economists in the 1990s?”, American Economic Review, vol. 82, no. 2, pp. 203–209.
  • Arnott, R. 1995. “Time for Revisionism on Rent Control?”, Journal of Economic Perspectives, vol. 9, no. 1, pp. 99–120.
  • Bertheim, C. 1948. “Housing in France”, Land Economics, vol. 24, no. 1, pp. 49–62.
  • Bonneval, L. 2011. “Le contrôle des loyers empêche-t-il l’investissement dans l’immobilier ?”, Métropolitiques, 12 October. (Also available in English: “Does Rent Control Prevent Investment in Real Estate?”, Metropolitics, 8 February 2019.)
  • Bonneval, L. and Robert, F. 2013. L’Immeuble de rapport: l’immobilier entre gestion et speculation (Lyon 1860‑1990), Rennes: Presses Universitaires de Rennes.
  • Bonneval, L. and Robert, F. 2010. “Mesurer la rentabilité du placement immobilier ; le cas de l’immeuble de rapport à Lyon (1890‑1968)”, Histoire & Mesure, vol. XXV, no. 1, pp. 127–163.
  • Croizé, J.-C. 2009. Politique et configuration du logement en France (1900‑1980), PhD thesis in human and societal sciences, Université Paris‑Ouest Nanterre La Défense.
  • de Jouvenel, B. 1948. “No Vacancies”, reprinted in Institute of Economic Affairs. 1972. Verdict on Rent Control, pp. 33–47.
  • Glaeser, E. and Luttmer, E. 2003. “The Misallocation of Housing under Rent Control”, American Economic Review, vol. 93, no. 4, pp. 1027–1046.
  • Hirsch, A. 1984. “Le Logement”, in A. Sauvy (ed.), Histoire économique de la France entre les deux guerres, vol. II, Paris: Economica, pp. 262–294.
  • Munch, J. and Svarer, M. 2002. “Rent Control and Tenancy Duration”, Journal of Urban Economics, vol. 52, no. 3, pp. 542–560.
  • Newsome, B. 2009. French Urban Planning 1940–1968, New York: Peter Lang.
  • Piketty, T. 2003. “Income Inequality in France, 1901–1998”, Journal of Political Economy, vol. 111, no. 5, pp. 1004–1042.
  • Simonnet, F., Gallais-Hamonno, G. and Arbulu, P. 1998. “Un siècle de placement immobilier. L’exemple de la fourmi immobilière”, Journal de la Société française de statistique, vol. 139, no. 2, pp. 95–135.
  • Sims, D. 2007. “Out of Control”, Journal of Urban Economics, vol. 51, no. 1, pp. 129–152.
  • Voldman, D. 2016. Locataires et Propriétaires. Une histoire française, Paris: Payot.
  • Wasmer, E. and Mayer, T. 2009. “Is There a Consensus Among French Economists?”, MPRA Paper 18584, University Library of Munich.
  • Willis, J. 1950. “A Short History of Rent Control Laws”, Cornell Law Review, vol. 36, no. 1, pp. 54–94.


[1See Paul Samuelson, Economics (8th ed., 1970, p. 372), and “Samuelson on Economics and Behavior” on PBS NewsHour (December 25, 2009), available online at: (in which Samuelson makes the exaggerated claim, without any supporting reference, that “the whole country of France had no residential building between World War I and World War II, primarily for the reason of permanent rent controls”).

[2The French law of 1948, today famous for the low rents it maintains in the few apartments to which it still applies, was primarily intended to permit increases in rents that had been frozen since 1914.

[3Willis (1950, p. 82), citing “La Crise du Logement et la Legislation des Loyers”, Notes documentaires et Études, no. 1028, Paris: Direction de la Documentation et de la Diffusion, p. 4.

[4Moreover, Bonneval’s calculations overestimate a building purchaser’s profits. He adds up the real value of rents due, instead of discounting that cash flow to take into account the time value of money. Money received in 1920 is more valuable than money received in 1938 because the recipient is able to invest it. The argument that Bonneval and Robert (2010, p. 156) make against the standard practice of calculating an internal rate of return does not justify their failure to apply a discount rate.

[5Bonneval (2011, p. 4): “building prices adjusted to [changes in] rents.”

[6See Bonneval and Robert (2013, figure on p. 153), which draws on Jacques Friggit’s studies in Paris and the authors’ own findings in Lyon.

[7Piketty (2003, p. 1020) confirms a 90% drop between 1913 and 1950.

[8See Sims (2007, pp. 143–144), who finds that rent-controlled dwellings were more likely to be poorly painted and plastered.

[9It is difficult to compare changes in housing quality over time. Nonetheless, according to Études et Conjoncture, the number of habitations vétustes (dilapidated dwellings) in France increased from 150,000 in 1911 to 2,800,000 in 1939. See Hirsch (1984, p. 283), citing “Le problème du logement”, Études et Conjoncture, “Union française” series (1947). See also Bertheim (1948, pp. 50–51), who describes the deplorable condition of French housing in the late 1940s.

[10Bonneval and Robert (2013, pp. 120–122). This lack of maintenance continued for several years after 1948 because the reform law enacted in that year generally did not allow rents to rise to market rates, which would have created stronger landlord incentives to maintain their buildings.

[11Because the construction of apartment buildings was so rarely profitable during the interwar period, many of the new dwellings that were built—indeed, perhaps a majority of them—were individual houses that households built for their own occupation (Hirsch 1984, p. 268).

[12Croizé (2009, vol. 2, pp. 100–109) describes uncertainties about the data that exist.

[13Croizé (2009, vol. 2, p. 101), citing INSEE (1980), Tableaux de l’économie française, p. 47.

[14See reference in the next footnote.

[15Newsome (2009, p. 27), citing “La reconstruction en France”, Documents économiques (1947, no. 31 p. 75). Other factors of course influenced these statistics. The population of France remained stable between 1914 and 1939, while the populations of both Britain and Germany were growing.

[16See the sources that Bonneval and Robert (2013) cite in their footnotes on pp. 16–17.

[17See, for example, Munch and Svarer (2002).

[18Arnott (1995, p. 102) offers a partial defense of mild “second-generation” rent controls.

To quote this article :

Robert C. Ellickson & David Le Bris, « The Damage Inflicted by French Rent Controls between 1914 and 1948 », Metropolitics, 26 February 2019. URL :
© - Toute reproduction interdite sans autorisation explicite de la rédaction