Communism, Embargo, and the Cuban Economy

Bastos, João Pedro, Vincent Geloso, and Jamie Bologna Pavlik, 2025, “The Forsaken Road: Reassessing Living Standards Following the Cuban Revolution and the American Embargo,” SSRN Working Paper no. 5235912, April 29.

How did the Cuban people fare economically under communism? According to government statistics, they benefited: Just prior to the 1959 Revolution, per-capita gross domestic product was just over $3,000 (measured in 2011 Geary–Khamis dollars, which allows for cross-country and cross-time comparisons), and it grew to just under $5,000 in 1989 when the Soviet Union fell, ending Moscow’s financial support for the island. Further, supporters of the Revolution say, per-capita GDP would have been much higher if not for the US embargo of Cuba.

But there are serious questions about these numbers’ reliability. There is broad consensus that Cuban authorities manipulate the economic data. Even if they are in earnest, Cuba has changed its national accounting system twice since the Revolution, making it difficult to compare data from different periods. Perhaps most important, prices in a centrally planned economy have little relationship to consumer valuation and scarcity.

Given those problems, researchers have developed alternative measures of Cuba’s post-Revolution economic performance. One is Queens College (New York) economic historian John Devereux’s series using sectoral output indices compiled by non-Havana sources. According to Devereux, Cuba’s per-capita GDP in 1989 was only about $3,450.

In this working paper, Bastos et al. build on Devereux’s work by comparing Cuba’s actual economic performance to how they project it would have performed if there were no Revolution. The authors construct a counterfactual non-communist “Cuba” from a weighted blend of similar Latin American economies, and they determine that Cuba’s actual 1989 per-capita GDP was 44.3 percent below the counterfactual. Things could have been worse: They project that if not for the Soviet subsidies (much of which went to Cuba’s sugar industry), the gap would have been 55.4 percent.

Is the US embargo to blame for this reversal of fortune? It would seem not, because Cuba found other trading partners. The authors’ figures suggest Cuban GDP per capita likely would have been only about 3 percent higher in the absence of the embargo, a percentage that is not statistically significant. To steelman this idea, they use two highly charitable scenarios based only on Cuba’s trade with the United States to inflate the embargo’s apparent cost, and even under those assumptions, estimated GDP per capita would have been only 7.6–10.6 percent higher without the embargo.

In short, the Cuban economy badly underperformed its potential in the decades after the Revolution and, if Soviet subsidies are excluded, it essentially stagnated. This shortfall was the result of communism itself, not the US embargo. A country that in the 1950s ranked among the wealthiest in Latin America and rivaled parts of Europe on living-standard indicators became a poor country by 1989.—TAF

Lighthouses, Protectionism, and Federalism

Geloso, Vincent, 2025, “Why Nationalize the Production of Public Goods?” SSRN Working Paper no. 5160519, April 22.

Econ 101 students learn that the free-rider problem leads markets to underprovide public goods, and government provision is justified to correct that shortfall. More advanced students read Ronald Coase’s “The Lighthouse in Economics” and learn that private and local-public provision of public goods often works well, competitively, and without state monopoly. Yet, centralized governments often take over provision of those goods. Why?

In this working paper, Geloso argues that governments nationalize public goods not because provision was failing when left to private or local control, but because public provision lets politicians extract rents in unrelated markets. Like Coase, Geloso draws on lighthouses to make his argument: in this case, American lighthouses in the Early Republic period (1789–1815). One seemingly arcane issue that was hotly debated by the Constitutional Convention was tonnage duties, the fees charged to ships based on their cargo capacity; states used that revenue (at least in part) to finance lighthouses.

The Constitution’s Article I, Section 10, stripped states of the authority to assess those duties, handing it to the federal government. That ended the interstate competition that had kept tonnage rates low and nondiscriminatory between foreign and domestic vessels. Under federal control, discriminatory rates against foreign ships jumped as high as 25 times that of domestic ships, creating a hidden tariff of 0.5–2.7 percent of import value. As a result, the foreign share of tonnage entering American ports fell from roughly 40 percent before federalization to under 10 percent within about a decade.

Once states lost the tonnage revenue, they had little reason to resist ceding lighthouse administration to Washington. Federal control then became a durable patronage tool: Customs collectors appointed lighthouse keepers and steered construction and supply contracts to political loyalists, and both Federalist and Republican presidencies reorganized the service to reward their own coalitions when power changed hands.

Geloso does not claim that nationalization reduced the quality of lighthouse service. Assessing whether federal takeover was a net social improvement would require comparing the safety gains from expanded lighthouse coverage under federal control against the deadweight losses from the protectionist tonnage duties—that is, weighing shipwrecks against lower trade volumes. So far, no research has tackled that question; instead, scholars typically see the many lighthouses built by Uncle Sam as an unalloyed good.—TAF

Oil and Gas Supply in the United States

Prest, Brian C., 2026, “Where Does the Marginal Methane Molecule Come From? Implications of LNG Exports for US Natural Gas Supply and Methane Emissions,” SSRN Working Paper no. 6671473, April.

In a 2021 post on the Cato-at-Liberty blog, I described important work by economists at Resources for the Future (RFF) that explained how the fracking revolution in oil and gas production has increased the responsiveness of oil and gas supply to price. This paper updates that work.

The effect of fracking on US production has been dramatic. In 2015, the United States imported 5 million barrels per day (MMbbl/​d) of crude oil and petroleum products on net; by the end of 2025 it was a net exporter of 3 MMbbl/​d. In 2023, the United States surpassed Australia and Qatar to become the world’s largest liquefied natural gas (LNG) exporter, shipping 12 billion cubic feet per day (bcf/​d). By 2025, US LNG exports were 15 bcf/​d and could be double that by 2029.

A permanent 10 percent increase in the price of oil increases supply by 6.1 percent. For comparison, in 2014, RFF economists assumed an oil supply response of only 4 percent, a third smaller.—PVD

SSDI

Deshpande, Manasi, Maxwell Kellogg, Magne Mogstad, et al., 2026, “Explaining the Historical Rise and Recent Decline in Social Security Disability Insurance Enrollment,” NBER Working Paper no. 35300, June.

Social Security Disability Insurance (SSDI) expenditures increased rapidly in the 1990s and 2000s. Regulation published articles and a book review on SSDI in the Fall 2011, Spring 2012, and Spring 2013 issues. The different authors agreed the expenditure growth was unsustainable, but they disagreed on whether statutory reform was required or if changes in the behavior of administrative law judges in the adjudication of benefit appeals would be sufficient to rein in this growth.

In the following decade, something not only reined in that growth, but reversed it. Unlike expenditure patterns in almost all other social welfare programs, expenditures on SSDI have declined since 2013. What explains the decline?

According to these authors, the largest factor (46–57 percent) in the decline is lower application rates, especially from younger men. Improved employment opportunities for less-skilled men explain nearly all the decline in SSDI application rates for men. In addition, 30–43 percent of the decline stems from lower award rates among those who apply because of administrative law judge reform.

The authors find no support for other possible explanations. Social Security field office closings were not a factor, and measures of hassle costs—like long wait and processing times—fell. There were no significant expansions of food or cash assistance programs or unemployment insurance. There is no relationship across states between health insurance increases and SSDI application declines. Veterans’ disability compensation did increase, but veterans contribute less than proportionally to the decline in SSDI applications. And population aging would have been expected to increase SSDI because of the shift to the highest-SSDI-receipt ages just below the full retirement age.—PVD

Employment Subsidies

Jain, Manisha, Corina Mommaerts, and Jeffrey Weaver, 2026, “The Limits of Targeted Hiring Subsidies: Evidence from the Work Opportunity Tax Credit,” NBER Working Paper no. 35229, May.

The Work Opportunity Tax Credit (WOTC) allows firms that hire applicants from designated disadvantaged groups—such as recipients of means-tested benefits or individuals with felony convictions—to claim tax credits for up to 40 percent of first-year wages, subject to a maximum credit of $2,400 per worker. The WOTC is a large program: Some 2.5 million new hires were certified to have their wages subsidized by the WOTC in 2022, which is more than twice the number of workers earning at or below the federal minimum wage, a policy that has attracted considerably more academic and policy attention.

The WOTC costs over $1.5 billion annually. Food assistance (SNAP) recipients are the largest category of beneficiaries, comprising two-thirds of WOTC certifications in recent years. The next largest, the long-term unemployed, comprised only 6.8 percent of WOTC-subsidized hires in 2023.

This paper asks whether the WOTC increased employment and earnings among SNAP recipients in Wisconsin from the late 1990s through the late 2010s. The research design exploits a major WOTC expansion in 2007 that increased the maximum age eligibility for SNAP recipients from 24 to 39. The authors compare labor market outcomes for SNAP recipients just below and above ages 25 and 40 around 2007.

The explanation for the findings is that firms rarely know whether a particular job applicant is subsidy-eligible. Only 20 percent of the firms that receive WOTC subsidies collect information during the hiring process that would allow them to identify WOTC-eligible applicants. Firms fear claims of age discrimination and thus do not ask age.

The WOTC program authorization expired at the end of 2025. This has occurred before; Congress has always renewed the program retroactively. A bipartisan group of nine senators is now co-sponsoring a bill to extend and expand the program, more than tripling its annual cost. That should not happen given the findings of this paper.—PVD