Recently there have been several estimates of Venezuelan economic assistance to Cuba — for example by Lopez (2012) and Mesa-Lago (2013). My latest estimates suggest that payments from Venezuela increased rapidly during the first decade of the XXI century and peaked at almost 19% of Cuba’s GDP in 2009. They declined over the following two years but remained quite large: I estimate Venezuelan assistance in 2011 (the last year for which the required data are available) at just over $7 billion, or 11% of Cuba’s GDP.2 These numbers are large, and they have invited comparisons with Soviet assistance to Cuba in the late 1980s. It has been implied that the adverse effect on Cuba’s real GDP of ending Venezuelan aid would be similar in size to the devastating impact of the elimination of Soviet aid in 1990. This is almost certainly wrong.
In the 1980s Cuba received large-scale assistance from the Soviet Union in the form of transfers and subsidies on Cuban oil imports and nickel and sugar exports. Soviet inflows averaged almost 23% of Cuba’s GDP in 1985–1988; they were sharply reduced in 1989 and eliminated in 1990. The result was dramatic: Cuban GDP plunged by about 40% from 1990 to 1993. Consumption, wages, investment and imports also collapsed, and large-scale shortages and power outages added to the suffering of the population. Only in 1994 did the economy begin to recover.
Early in the first decade of the XXI century, Cuba began to receive substantial assistance from Venezuela in the form of favorable financing terms for oil imports, transfers, and payments for the work of Cuban doctors, teachers and other professionals in Venezuela. There is no doubt that the termination of Venezuelan assistance would have a serious impact on the Cuban economy. However, there are three strong reasons to believe that the resulting damage would be small in comparison with the cataclysmic consequences of the withdrawal of Soviet aid more than two decades earlier.
- The first reason is that inflows from Venezuela in 2011 were less than one half of Soviet inflows in 1998. Therefore, Cuba would now experience a lower reduction in external saving, and therefore lower reductions in domestic investment, than in the early 1990s.
- A technological shock. The breakdown of relations with the USSR resulted in a massive change in the geographic structure of Cuba’s foreign trade (see Figure 1). Imports from the USSR (and then from Russia) almost disappeared, and the stock of usable capital was severely curtailed as it became virtually impossible to import machinery, equipment and parts from Russia and its former satellites. This was not just because the drop in external saving reduced the funds available to finance investment. Even if foreign saving had been available (or had been replaced by domestic saving) it would still have been almost impossible to import capital goods and parts consistent with Cuba’s existing Soviet technology. The problem was compounded by the departure of Russian and East-European advisors. The rupture with the USSR (and then more radically with Russia) made repair and maintenance a veritable nightmare for Cuban enterprises and made much of the existing capital stock useless. Not much could be done about that in the short term because Soviet technology was deeply embedded in the Cuban economy. Nothing of that sort is likely to happen now because Venezuelan technology does not have such a predominant role in Cuba and, more to the point, because there is no such thing as Venezuelan technology. 3
- Mistaken macroeconomic policies. Cuba reacted to the “Soviet shock” by foolishly attempting to replace real Soviet subsides by domestic budgetary subsidies financed by monetary expansion. Since most domestic prices were controlled, the surge in the money supply created a massive monetary overhang and a fall in household consumption through forced saving and widespread rationing, and the meager saving of household were vaporized by the inflation tax. Of course, this ill-advised policy did not erase the enterprise losses created by the collapse of foreign saving and domestic investment. Only in 1994 did the situation begin to improve after subsidies for enterprise losses were cut, the economy was aggressively stabilized (and modestly liberalized), and Soviet-style technology was gradually replaced.
Figure 1. Cuba: Geographic Structure of Imports, 1989–1998 (in percent of total imports)
The dramatic deterioration of the policy environment in the immediate post-Soviet area can be seen by comparing the first two columns of Table 1.
Money growth, the fiscal deficit, and subsidies for enterprise losses surged; the gap between market and official exchange rates widened dramatically; and non-state (private and cooperative) employment remained extremely low.
Table 1. Cuba: Selected Indicators of Macroeconomic Instability, Efficiency and Dependence (In percent of GDP unless otherwise noted)
The present situation is very different from the one prevailing in the early 1990s: a comparison of the second and third columns in Table 1 shows a sharp improvement in all macro- and micro-economic indicators.4 In particular, there were sharp reductions in subsidies for enterprise losses, the fiscal deficit, the ratio of money to GDP, and the share of state employment. Even though statistical information is not available, there is evidence that some prices have been liberalized and that others have been adjusted more frequently.5 Private sales and purchases of houses, cars and some consumer durables by private agents have been authorized, and the exchange rate gap is now lower, although the misalignment remains huge.
In one important respect, the present situation is similar to the one prevailing in the late 1980s, beforethe elimination of Soviet aid: the geographic distribution of trade has changed dramatically and it is, again, highly concentrated in a single country — this time Venezuela, which has become by far Cuba’s main trading partner (Figure 2). The dependence on foreign aid is again quite large, but lower than it was towards the end of the Soviet era. Otherwise, the situation is now quite different. Comparing the first and third columns of Table 1 shows that the fiscal deficit is lower (in spite of an unusual surge in 2008); subsidies for enterprise losses are much lower; non-state employment is much higher; and dependence on food and petroleum imports has been reduced.
In sum, while the political future is unpredictable, it is doubtful that President Raúl Castro’s administration will reverse course at this juncture and destroy the progress made in stabilization and reform by repeating the policy mistakes of the early 1990s.
Figure 2. Cuba: Geographic Structure of Imports, 2005–2011 (in percent of total imports)
Table 2. Cuba: Contribution of the Explanatory Variables to the Change in Real GDP, 1991–1993
Table 3. Impact of the Elimination Foreign Assistance on Cuba’s GDP (Percent changes during relevant period(a))
A MODEL-BASED APPROACH TO ESTIMATING CUTS IN FOREIGN ASSISTANCE
This section attempts to quantify and compare the economic events of the early 1990s and those that might occur following a hypothetical cancellation of current agreements with Venezuela.
Kurazh zakonchen: The End of the Party. In an earlier study, I examined the end of Soviet assistance to Cuba and its consequences using a production function relating real GDP to the usual variables (capital stock and employment). In addition, the regressions included proxies for technological disruptions and for the policy reaction of the Cuban authorities before 1994. (See Hernández-Catá, 2013 and Annex 1 of this paper.) On the basis of that study, this section tries to account for the contraction of output in the period 1991–93 (i.e., from 1990 to 1993). Table 2 shows the rate of change of the various explanatory variables during that period (column 1) and the corresponding regressions coefficients (column 2). Column 3 shows the estimated contribution of each explanatory variable to the decline in output, obtained by multiplying columns 1 and 2. The estimated total impact on real GDP, nearly 38%, is the sum of contributions of all the explanatory variables.
In column 1 of Table 3 the output effects are listed under three main headings: (i) the decline due to direct macroeconomic changes; (ii) the technological disruption effect; and (iii) the effects of the misguided Cuban policy reaction. Each of these three variables explains approximately one third of the total estimated drop in production from 1990 to 1994.
As part of the direct macroeconomic effect, the fall in the utilized capital stock (ignoring the obsolescence resulting from the inability to obtain Soviet equipment and parts) accounts for 13.5% of the total decline, and the deterioration of the terms of trade for roughly 9%. The decline in employment explains only 3% of the total decline in production — a small contribution reflecting the government’s propensity to absorb high unemployment by keeping unproductive workers in the state payroll.
Several aspects of Table 3 should be mentioned. First, the total actual decline in output from 1991 to 1993 (just under 40%) is much larger than the initial ratio of Soviet assistance to Cuban GDP, illustrating the fallacy of evaluating the consequences of cutting aid on the basis of its initial magnitude. The contributions of technology and policy are correspondingly very large. Second, in the absence of the mistaken domestic policy reaction to the crisis, the estimated contraction of production would have been 26% instead of 38%.
Se acabó lo que se daba: The Effects of a Hypothetical Elimination of Venezuelan Assistance. Here, the problem is a little more complicated, because data for the explanatory variables is not available for the relevant (future) period. As an alternative we specify a priori the changes in output that would result from a reduction in external saving (using data for 2011), and then apply the regression coefficients shown in Table 2 to estimate the contribution of each explanatory variables. (This is spelled out in Annex 1.)
The estimates provided in Table 3 (columns 2 and 3) indicate that the contraction of output due to a hypothetical post-Venezuelan shock is likely to be in the range of 7% to 10%, considerably less than the estimated post-Soviet impact of almost 38%. This is not only because the ratio of foreign assistance to GDP was much larger at the end of the Soviet period than in 2011 (the starting point for the post-Venezuelan exercise). It is mainly because the technological and domestic policy factors are now assumed not to play a role. The first assumption simply reflects the historical impossibility of a Venezuelan “technological” effect.
The second assumption reflects the belief that the misguided macro policies of the post-Soviet era will not be implemented by the Raúl Castro administration in response to an end of Venezuelan assistance. This assumption seems reasonable at this stage given the present economic situation in Cuba and the reforms that have already been implemented. Nevertheless, a cursory examination of the implications of alternative policy scenarios may still be of interest. As an illustrative exercise, the third column of Table 3 assumes that the Cuban government would extend budgetary subsidies to state enterprises in a (futile) attempt to avoid the drop in investment that would otherwise result from the withdrawal of Venezuelan payments. These subsidies would be financed by monetary expansion and therefore give rise to a monetary overhang, forced saving and inflationary taxation that would contribute an estimated 1.7 percentage point drop in real GDP (Annex 1).
A comparison of the first two columns of Table 3 shows that the negative employment effect is the same in the post–Venezuelan case as in the post-Soviet case, in spite of the much larger contraction of real GDP in the latter case. The reason for this result is that the Cuban government in the early 1990s was very reluctant to accept a large increase in open unemployment. However, there are clear indications that the government’s propensity to absorb redundant workers is now considerably lower than in the past — as evidenced by the dismissal in 2011 of about 300,000 state employees that eventually found their way into cooperatives or in the small private services sector. Thus, the employment effect is higher relative to the output effect in column 2 than in column 1, because it is based on a higher income-elasticity. (This is explained more fully in Annex 1.)
By contrast, the alternative bad-policies scenario of column 3 assumes that the government returns to its earlier policy of resisting open unemployment, and that the subsidies to cover the corresponding rise in the wage bill are financed by money growth. The result, calculated by multiplying the regression coefficient of the monetary overhang variable (from Table 2) by the required increase in domestic subsidies, is to lower output by an additional 0.9 percentage point. In the third column of Table 3, the line labeled “policy mistakes” is the sum of the effects of subsidies to “protect” investment and employment, respectively.
The analysis presented in this paper indicates that a complete cancellation of Venezuelan assistance to Cuba would cause considerably less damage than the elimination of Soviet assistance in the early 1990s, with the fall in real GDP estimated at somewhere between 7% and 10%, compared to 38% after the breakdown of Cuban/Soviet relations. Moreover, if the Cuban government were to avoid the policies of subsidization and inflationary finance pursued in the post-Soviet period, the post-Venezuelan contraction would be at the lower end of the range or approximately 7%.6
This is still a lot, however. To be sure, the danger of a sudden elimination of aid inflows has diminished considerably since the Venezuelan election of April 2013. Nevertheless, the prospect of a more gradual reduction in aid remains likely given Venezuela’s economic difficulties. In that case, the effect would be a reduction in the growth of the Cuban economy spread over several years, rather than a sudden contraction of output. Furthermore, current efforts to obtain financing at non-market terms from other countries, like Algeria, Angola and Brazil, would, if successful, diminish the magnitude of the shock. But it would perpetuate dependence and delay the needed adjustment.
The only way to diminish the pain of reduced income and consumption would be a decisive effort to expand Cuba’s productive capacity by intensifying the reform process. The list of required actions is familiar to all: liberalize prices, unify the exchange rate system, dismantle exchange and trade controls, stop the bureaucratic interference with non-state agricultural producers, continue efforts to downsize employment in the state sector, and increase substantially the list of activities opened to the private sector, including (why not?) doctors, nurses, teachers and athletes. Private clinics and schools would pop up, consultancy services would flourish, and the baseball winter leagues would come back to life.
Karl Marx (1852) credited Hegel with the idea that history repeats itself twice. Unfortunately for him, he added: “the first time as a tragedy, the second time as a farce.” This is not necessarily true. Often the second time is also a tragedy, as when the West gave Eastern Europe to Stalin at Yalta, less than a decade after giving it to Hitler in Munich. And why couldn’t the second time be an epiphany? Cuba’s rulers now have a historic opportunity to allow people to improve their own standard of living, and to stop wasting resources to keep the faded and sinister red banner afloat. Without a doubt, history will absolve them if they take that chance. And then, perhaps, Cuba will be allowed to replace its politically inspired dependence on doubtful friends with free, mutually beneficial trade with all nations.
Annex 1: The model
Annex 2: Estimating Venezuelan Assistance to Cuba
The estimate of $7 billion, or 11% of Cuban GDP, for Venezuelan payments to Cuba in 2011 is an average of two independent estimates. The first is derived by subtracting “other transfers to enterprises” (‘transferencias al sector empresarial, otros’) from “other non-tax revenue” (‘otros ingresos no tributarios’). This operation helps to adjust the budget numbers to national income accounts basis — the appropriate concept to estimate the impact on GDP of eliminating Venezuelan aid. All variables are from (ONE 2012), State Budget Performance table, according to which “other non-tax revenue” includes “price differentials in foreign trade and net foreign income.”
The second estimate is calculated by adding central government outlays corresponding to services of Cuban medical and teaching personnel abroad to a rough estimate of investment projects financed by Venezuela. A third estimate (not used in this paper) was derived by replacing central government spending on health and education by exports of services (balance of payments basis) minus tourism and transportation. The resulting series is highly correlated with the first estimate, but displays somewhat larger errors and omissions.
It is important to note that the value of oil imports from Venezuela is not included in the definition of Sf because these imports, contrary to a common misunderstanding, are valued at world market prices and will need to be replaced by imports from other countries (also at world market prices) after the cancellation of the program with Venezuela. Unless the new suppliers agree to Venezuelan terms, new oil imports would have to be financed at a higher interest. (Oil imports from Venezuela now carry an interest rate of 1%, but they do not involve an interest payments at present because of the grace period specified in the agreement.) Oil imports amounted to $3.5 billion in 2011, excluding oil imports to be refined in Cuba and re-exported. This estimate is based on the methodology provided by Piñón and Benjamin-Alvarado (2010), explained in the footnotes to Table 2–5 of their article.
1. I would like to thank Luis Luis for extremely valuable comments on an earlier draft.
2. The derivation of these estimates is explained in Annex 2.
3. If anything, technology transfers go the other way around. Hernandez (2013) reports that Albet (a Cuban software company) has signed a contract with the Venezuelan government to provide and administer electronic identification documents , maintain a civilian registry, centralize information on the country’s prison, security, emergency and hospital systems, and monitor oil rigs managed by the Venezuelan oil company PDVSA.
4. Subsidies for price differentials are the only exception, highlighting the continued difficulties in eliminating the rationing system.
5. The relative price of electricity and gas increased by 54% in 2006, while the relative deflator for the sugar industry more than doubled in 2010. Relative prices are measured by the ratio of own to total GDP deflators in the corresponding sector.
6. There may well be other adverse effects down the line. For example, the shift from Venezuela to other oil suppliers may involve a shift from long-term, subsidized debt with a generous grace period to unsubsidized short-term borrowing with shorter grace period, if any. This would occur to the extent that Cuba fails to persuade new oil exporters to duplicate the term now provided by Venezuela. In addition, unless the alternative suppliers continue to provide additional quantities of oil to be refined in Cuba and then re-exported, Cuba could suffer a loss of output and jobs at the Cienfuegos refinery.
7. An alternative that yields exactly the same result is to solve the first difference equation for the change in output, using the formula Δy = (α ΔSf / K + Δx)/φ, where φ = 1–αϕ η – (1–α)η. However, this method does not allow the separate calculation of the various contributions.