The multiple exchange rate system that has existed in Cuba since 1995 is a remarkable case of self-inflicted injury. It has introduced severe distortions, thus lowering productivity and incomes. It has discriminated severely against exporting firms, and has provided a huge subsidy to importers, thus harming the country’s trade balance. The authorities have announced their intention to liberalize the exchange system, but the process has been exasperatingly slow.
During the post-Soviet tragedy in the early 1990s, the explosion of money-financed fiscal deficits resulted in a huge depreciation of the peso in the parallel market (Figure 1). In 1995, as the economy was being aggressively stabilized, the government decriminalized the possession of the U.S. dollar and allowed households to buy and sell foreign currency in exchange houses (CADECAs), where the dollar/peso rate was allowed to float. However, the official exchange rate of one Cuban peso per U.S. dollar continued to apply to state enterprises and to the conversion of official statistics.
The dual exchange rate was born. The 1995 measures were a double-edged sword: the recognition of a freely floating, market-determined exchange rate for households was a welcome initiative, but the maintenance of an administratively fixed and highly over-valued rate for enterprises was a major mistake that created many of the problems that have haunted the Cuban economy since then.
A PARENTHESIS: CUC VS CUP
Another currency, the so-called “convertible peso” or CUC, was introduced in 1995. Like the enterpriseonly Cuban peso (CUP), it was valued at 1 CUC per U.S. dollar. So we now had a multiple, rather than a dual, exchange rate system. But the CUC differed from the CUP in that its convertibility was backed by official reserves. The Central Bank of Cuba would buy and sell foreign exchange reserves so as to maintain the value of the CUC at par with the U.S. dollar. In a sense the device was similar to the ill-fated Argentinian currency board arrangement—the Arreglo Monetario. Like its southern cousin the CUC probably would have ended in spectacular disaster had the Central Bank of Cuba not ended the charade by eliminating the convertibility of the CUC in 2004, after using up a large chunk of Cuba’s foreign exchange reserves. 1 The announcement of a plan to unify the exchange system in 2013 signaled the demise of the CUC, and since then there have been indications of large scale central bank purchases of CUCs against CUPs (see Vidal Alejandro and Hernández-Catá, 2016). 2 A merciful end for a bad idea.
It is often asserted that the essence of the Cuban multiple rate system is the existence of two currencies: the inconvertible CUP and the “convertible” CUC, but this misses the point. From an analytical standpoint, the CUC is a sideshow that can be safely ignored. The most serious and harmful duality has not been between CUC and CUP, but between the two types of CUPs: the one that applies to households (now 24–25 CUPs per US dollar 3); and one that applies to enterprises (1 CUP per US dollar). 4
A LEGACY: OF TROUBLES
The dual (or more accurately, the multiple) exchange rate scheme, in combination with the system of price controls, has led to widespread distortions that have damaged productivity for years. It has depressed Cu-ba’s strikingly weak merchandise export sector and has come under intense criticism for many years, from both sides of the strait of Florida.
The system introduces various forms of discrimination by having households buy and sell dollars at the CADECA rate, while enterprises buy and sell at the 1 CUP/dollar rate. Specifically, it discriminates against households wishing to buy dollars, and against exporters and providers of tourist services wishing to sell dollars. It discriminates in favor of import-competing industries and imports—notably imports of oil, thus encouraging energy consumption. Over time, therefore, the system has contributed to a pronounced deterioration of Cuba’s merchandise trade balance. An overall balance of payments crisis has been averted only because of the steep rise in remittances 5 and, until recently, of exports of professional services provided in exchange for Venezuelan oil.
In addition, the system has seriously complicated the authorities’ task of analyzing and interpreting economic developments by using an overvalued exchange rate of 1 peso per US$ as a conversion factor for official statistics
Another problematic aspect of the existing multiple rate system is that it raises revenue for the government. Workers in the tourist industry and Cuban professionals working in Venezuela, Bolivia and other places earn dollars, but the Cuban government takes a cut and pays the workers in CUPs, thus imposing a huge tax. In addition, there is a 10% tax on exchanges of CUPs for U.S. dollars. Presumably, this scheme will disappear with unification and, unless replaced by an explicit tax, alternative sources will have to be sought or, better yet, spending will have to be reduced.
If the system is so damaging, why has it taken so long to get rid of it?
Reform has been opposed by those who benefit from the old system, and by those who oppose any change in the direction of liberalization for ideological reason. Reform also has been discouraged by fears, propagated in Havana as well as in Miami and Puerto Rico, based on four myths that can be quickly debunked.
Myth #1. Households will be hurt by the increase in the cost of imported goods that will result from the depreciation of the CUP. No. This is a misunderstanding. The CADECA rate that applies to households is now probably close to a free market rate. Therefore, it will probably not be significantly affected by unification and liberalization
Myth # 2. It is necessary to liberalize the domestic economy before reforming the exchange rate system. The dog is chasing its tail: this is a recipe for inaction.
Myth # 3. The creation of a free and unified exchange market would force the government and the enterprises to assume the exchange risk. But changing the exchange system will neither abolish nor generate risk. A fixed exchange rate does not eliminate the risk of a devaluation, as proved by the experience of Mexico, Argentina, and many other countries in the 1990s. Since the one-to-one CUP/$ exchange rate that applies to enterprises is strongly overvalued, it creates the risk of a massive future devaluation, against which it is very difficult to hedge. In the future, Cuban enterprises should be allowed to deal with exchange risk through forward and futures markets, which of course requires a little bit of institutional work. The key point here is that the inevitable devaluation of the enterprise-only CUP will stimulate exports and tourism and contain imports. That’s what the doctor ordered and what the Cuban balance of payments requires.
Myth # 4. Foreign investors and exporters will not accept payment in Cuban pesos. Why not if the peso is convertible and is backed by disciplined monetary and fiscal policies? There is no reason why this cannot be achieved. After all, with a few exceptions, macroeconomic policies in Cuba have been responsible since 1994 and have delivered price stability.
A TRIUMPH OF INTELLIGENCE. BUT ARE WE REALLY ON OUR WAY OUT?
In 2013, the Cuban authorities finally decided to reform the exchange system: they announced measures that would lead to the unification of currencies and exchange rates. The plan envisaged a devaluation of the enterprise-only CUP, making the CUP the centerpiece of the new system, and retiring the CUC (“Official Note”, 2013). This was all entirely appropriate and, at that time I called the decision a triumph of intelligence. And that’s what it was. Cuba’s inefficient and discriminatory multiple exchange rate system was on its way out. But the remaining questions are how and when?
Unfortunately, the authorities decided to pursue a gradual and sectoral strategy. Gradual is self-explanatory: this would not be a Polish or Vietnamese-type operation but one that would take time to complete. Sectoral means that reforms of an experimental nature would be introduced at differential speeds in various sectors of the economy. For example, already the value of the CUP for transactions between state hotels and restaurants and agricultural cooperatives was devalued from 1 CUP per US$ to 7 CUPs/US$ in 2011, and to 10 CUPs per US$ in 2013 (Figure 2). The system now included a high multiplicity of exchange rates, but the process of unification appeared to be on its way. Unfortunately we have no information on what happened after 2013.
To summarize, the direct effects of the plan on sectoral earnings would be as follows: exporters will be better off, importers will be worse off, and there would be no substantial direct impact on households. 6 Over the long term these changes will raise productivity and improve the balance of payments. The devaluation also will have consequences for the net asset position of different actors. It will be unfavorable for those enterprises with foreign currency liabilities exceeding foreign currency assets, and favorable in the opposite case. As for households the impact should be small as the government has announced that it will maintain the value of peso-denominated saving accounts—one of the population’s main depositories of wealth. In contrast, the central government has a large stock of foreign currency-denominated debt and it will therefore suffer a strong negative effect.
THE PERILS OF GRADUALISM
I see two main problems with the approach that is being pursued.
Gradualism gives time to those who oppose the reforms to counter-attack and thwart, or at least delay, the process. 7 In the meantime, many of the distortions associated with the old system persist. The champions of gradualism stress that the move toward a proper conversion factor for official statistics will require careful preparation and training. But it has been four years since the start of the reform and adequate training cannot take much longer that!
The principal argument in favor of gradualism is that the profitability of import-competing enterprises will be eroded by the devaluation of the peso (in some cases dramatically) and they might collapse unless they are given time to adjust. However, the experience of the former Soviet Union suggests (a) that some of these firms will be unable to adjust and will have to be closed anyway; and (b) the former Soviet republics that liberalized rapidly achieved the fastest recoveries and the fastest growth of GDP after the reforms. De la Torre and Ize (2014) suggested providing temporary subsidies to the most seriously affected import-competing industries. This is an interesting idea, but “temporary” subsidies have a way of becoming permanent; the effort and time required to get rid of the subsidies granted in the immediate post-Soviet period should not be forgotten. Providing new subsidies at a time when the public finances will be under pressure to establish the credibility of macroeconomic policies in support of the new exchange system.
It could be argued that the gradual exchange rate changes made so far reduce the over-valuation of the peso in some sectors, and therefore they are steps in the right direction. But from a resource allocation perspective this is not necessarily so: the Lipsey-Lancaster theorem states that partial removal of distortions is not always a second-best.
Another danger of gradualism is speculation. Since the prospective devaluation of the enterprise-only CUP will probably be massive, a fortune could be made by buying dollars now, for at 1$/CUP (for example by over-invoicing imports) and holding these dollars abroad (or at home under the pillow) awaiting the time when the dollars could be sold around CUP24/dollar. Another mechanism would be for firms to (illegally) withdraw CUP deposits from Cuban banks, seek ways to exchange them for dollars at par, and keep the proceeds abroad until the expected depreciation of the CUP is complete. We cannot be sure that these things, which are of course illegal, are happening on a significant scale. But the evidence provided by Luis (2017) suggests that in recent years the Cuban balance of payments has shown large outflows of capital in the form of negative net errors and omissions. This could be countered by observing that data for peso-denominated household deposits do not indicate that speculative withdrawals have occurred, at least not on a large scale. But this is not a very good test because the published data include only household deposits and exclude enterprise deposits which are more likely to be involved in speculative activity. In any event, the longer the incentive is in place the more difficult it will be to avoid the temptation to take advantage of the existing huge exchange rate gap, particularly if the authorities; commitment to eliminate this gap is credible. 8
SOME UNSOLICITED ADVICE
The dangers inherent in the gradualist/sectoral approach to exchange rate unification in my view largely exceed any merits it might have. My preference would be to complete the reform as soon as possible—in other word to cut the cat’s tail in one installment and thus avoid prolonged anxiety. This would avoid the risk of speculation associated with gradualism. It would also avoid further slippages, and yield the benefits of improved resource allocation, export performance and external sustainability at an early stage. It could be argued that, given the magnitude of the gap between household-CUPs and enterpriseCUPs, an immediate move would cause a traumatic shock to the economy. That is not very convincing. After all, the equally huge depreciations in several countries of Eastern Europe and the former Soviet Union and in Vietnam went rather smoothly and did not produce any visible dislocations.
Early experimentation with markets might be helpful. For example, CADECAs should be authorized to let the household-CUP move freely within a prespecified band that might be expanded over time.
This would create a proto-foreign exchange market in which the central bank could conduct hard currency auctions, thus allowing participants to gain experience and provide an indication of the rate at which the enterprise-CUP might be expected to converge. The elimination of the 10% tax on dollar/peso conversions would also help by providing depth to the market.
When liberalization is complete, my advice is to resist the dangerous temptation to fix the value of the peso. Instead, just let it float and, if you still have some reserves left, lean against the wind à la Bank of Canada.
Hernández-Catá, Ernesto (2014). “Macroeconomic Policy in Cuba. History, Problems and the Agenda for the Future.” In Bryan Christiansen, editor, Economic Growth and Technological Change in Latin America. London: 2014.
Luis, Luis R. (2017). “The Capital Account of the Cuban Balance of Payments.” Forthcoming in Cuba in Transition, Vol. 27. Association for the Study of the Cuban Economy, Miami.
“Official Note and Economic and Social Policy Guideline No. 50 of the Party and the Revolution” (2013). Granma International, Havana (October 22).
De la Torre, Augusto and Alain Ize (2014). “La Unificación del Tipo de Cambio: El Caso Cubano”. Cuba in Transition, Vol. 24. Association for the Study of the Cuban Economy, Miami, FL.
Vidal Alejandro, Pavel (2013). “El tránsito hacia una sola moneda es clave para el éxito de la reforma en Cuba”. Entrevista a Pavel Vidal por Lenier Gonzáles Mederos. Espacio Laical. La Habana (Octubre).
Vidal Alejandro, Pavel and Ernesto Hernández-Catá (2016). “The Rise and Fall of Economic Growth in Cuba, 2015–2016. External Factors, Business Confidence and Domestic Policies. Cuba in Transition, Vol. 26. Association for the Study of the Cuban Economy, Miami.
- Cuba’s claims on financial institutions in the Bank for International Settlements (BIS) area fell by $1.3 billion in 2003, or more than half of Cuba’s net claims. BIS data do not cover all of Cuba’s foreign asset and liabilities because they do not includes positions vis-à-vis certain countries, notably Russia and China.
- This conclusion is based on an examination of the difference between the fiscal deficit and the change in the CUP-denominated broad money supply (M2A) net of net of the change in net foreign assets (proxied by the change in Cuba’s net claims on banks reporting to the BIS). In 2014–15 this difference became extremely large by historical standards.
- These refer to buying and selling rates.
- See Hernández-Catá (2014).
- The system does not penalize remittances which can be converted from dollars into to Cuban pesos at the CADECA rate.
- Of course the devaluation of the enterprise-only CUP would have indirect effects on the household sector. These effects would be ambiguous in the short term but should be unambiguously favorable in the medium to long term.
- The statement in the Official Note (2013) that subsidized retail prices for “people in need” will continue is also of concern. Logic and fairness (not ideology) requires that the price controls associated with these subsidies be eliminated and replaced by targeted fiscal transfers, at no cost to those truly in need. The point is to eliminate distortions, whether they result from the exchange rate system or from price controls.
- It could also be pointed out that Cuban banks offer dollar-denominated deposits; if you want to hold dollars that’s one way to do it legally. But as they say in Mexico, the only true dollar is the “verde” (the “green”). Residency matters in these things. It should also be noted that ONEI publishes data for deposits held by households, but not for those held by enterprises.