To the Editor:
Recent publication in Academic Questions of impressive graphs exhibiting a strong positive correlation between “State Ownership of the Economy, 1900-2022” and an overall “Academic Freedom Index,” as well as with five separate components of the latter variable are interpreted by the authors as demonstrating that socialism is the road to academic censorship (“Socialism: The Road to Academic Censorship,” Fall 2025)1 The paper offers a textbook case of the abuse and misinterpretation of statistics in support of a political argument.
First, as is well known, correlation of two variables does not establish causality. Iced-tea sales and drowning deaths may be highly correlated, but restricting sales of iced tea won’t prevent drownings. Both are driven by an independent factor, hot weather.
Stratifying on an independent variable provides an explanation for misleading associations. In the case of socialism and academic freedom, the obvious “hot weather” independent variable is the type of government, whether democratic or not. Dictators are more likely to own the means of production, and more likely to limit academic freedom, along with a host of other freedoms.
The authors argue that a causal relationship of socialism with restriction on academic freedom appears plausible because socialism 1) must restrict employment options for academics, both private and public, 2) limits the private wealth which provides “diverse funding sources” to academia, and 3) restricts control of necessary resources for expression of “public intellectualism.”
The first is simply absurd. If anything, critics of socialism often charge that government ownership is necessarily inefficient, and must lead to bloated payrolls, not restricted employment! It is simply implausible that academics would have fewer academic job options if a private industry is nationalized. The other two are conjectural and unlikely. Certainly, fascist states limit if not prevent wealthy individuals from supporting expression of views uncongenial to the rulers. Indeed, the only people likely to accumulate wealth in such a state would be those in agreement with its leaders.
And what has happened in democracies which have adopted or given up socialist policies? Did academic freedom in the United Kingdom diminish when the Labor Government after World War II nationalized the coal and steel industries, or introduced the national health service? Did it increase when Margaret Thatcher privatized many industries? Of course not.
Or consider the post office, which in almost all countries is government owned and operated. Privatizing the post office will not increase academic freedom. Nor would privatizing Amtrak, TVA, Fannie Mae, Freddie Mae, Ginnie Mae, The Bonneville Power administration, The Export-Import Bank, The FDIC, The Pension Benefit Guaranty Corporation, and the many local government owned transit systems, airports, utilities, or the state-owned liquor stores in Vermont and elsewhere increase academic freedom.
Peculiarly, the authors do not undertake a fine-grained analysis of factors which do restrict academic freedom in democracies. In our own country such constraints have usually been driven by religious, moral, and patriotic crusades against academics expressing unpopular opinions.
Indeed, one could make a strong argument that the fervor, depth, and extent of religious practice in a country is not only correlated with restraints on academic freedom but is causally connected to this outcome, but I will leave that to a more detailed analysis.
Ernest B. Hook
Professor Emeritus, School of Public Health, University of California, Berkeley
Daniel J. Smith Responds:
We appreciate Professor Hook’s response to our paper and welcome the opportunity to build upon and clarify our arguments and methodology. Most importantly, we did not intend to imply a causal claim between state ownership of the economy and academic freedom. Rather, our approach is historical and descriptive. Using long-run data that spans from 1900, we document broad empirical patterns not causal effects. The observed pattern, that socialist countries have been unable to maintain academic freedom, is noteworthy and a natural prelude to more in-depth causal analysis. Berggren and Bjørnskov (2024), for instance, provide a causal analysis using economic freedom and academic freedom.
As such, we do not argue that a given decrease in state ownership of the economy—that is, more socialism— produces a specific or measurable erosion of academic freedom, since we do not run causal tests. In related work using the same measure of state ownership of the economy, we document that sustained democratic institutions are not observed under high levels of state ownership or direct control of the means of production.2 While not all non-democratic regimes are socialist, economies that have reached a socialist stage are consistently observed to be non-democratic. Our argument on academic freedom can be seen as an extension of Hayek’s argument in The Road to Serfdom—academic freedom is historically incompatible with socialist economies just like other civil liberties.
Some confusion may result from the varying definitions of socialism. While we define socialism in the paper as state ownership or direct control of the means of production, we accept responsibility for failing to provide a detailed discussion of the V-Dem codebook for the State Ownership of the Economy variable. That variable ranges from 0 to 4 and measures the extent to which the state owns or directly controls valuable capital across sectors (0 is complete state ownership or control). Consistent with that definition, we classify scores between 0 and 1 as representing socialist economies, characterized by state control. Our analysis focuses on this range, as may be inferred from Table 1 in our original article. Responding directly to Professor Hook, the United Kingdom did maintain a high academic freedom index score during its nationalization and subsequent privatization. But this does not undermine our argument since they were never a socialist country; the United Kingdom’s lowest score was 2.3, a mixed economy.
Another benefit of this exchange is that we realized from Professor Hook’s comments that we did not properly clarify what we mean by “employment options.” The argument is not that nationalization necessarily reduces the total number of jobs, or even academic jobs. Our claim concerns the structure of employment, not its size. When a private industry is nationalized, academics may still face many jobs in the aggregate, but those jobs become more similar because the employer is now the same: the state. Competition among employers disappears, exit options shrink, bargaining power declines, and thus power becomes concentrated, narrowing the space for dissent.
In economics, this is referred to as monopsony. If academia was nationalized and all universities were government-owned, the state would be the dominant buyer of academic labor. In such a setting, the demand side of the market holds disproportionate power, making it easier to influence hiring, funding, and ultimately the boundaries of what is considered acceptable scholarly expression. The power to exclude enables the wielder of that power to enforce conformity. Thus, when employment is centralized under a single employer, scholars lose credible exit options, and academic freedom weakens accordingly.
Finally, we agree that other factors may restrict academic freedom in Western democracies, including religious, moral, or patriotic pressures. These are important topics and merit separate analysis. They are simply not the focus of this paper. What we aim to show is the historical relationship between the overall economic system and academic freedom, and we believe the results are compelling as such. The claim is narrow, structural, and historical. When the government controls the means of production and economic activity, one should not expect academic freedom to persist. That is the core argument of the paper.
We recognize that the presentation of long-run data can sometimes invite causal interpretations, and we appreciate the opportunity to clarify our intended scope.
Daniel J. Smith
Middle Tennessee State University
1 Gabriel F. Benzecry, Nicholas A. Reinarts, Daniel J. Smith, “Socialism: The Road to Academic Censorship,” Academic Questions 38, no. 3 (2025): 30-7.
2 G. Benzecry, Nicholas Reinarts, Daniel J. Smith, “You Have Nothing to Lose but Your Chains? Re-examining the Hayek-Friedman Hypothesis on the Relationship Between Capitalism and Political Freedom,” Public Choice 202 (2024): 529-556.
Photo by Andrew Seaman on Unsplash