Ethiopia’s experiment (continued, by someone actually knowledgable)

This is a guest post from CGD’s and IIE’s Arvind Subramanian.

Chris drew attention yesterday to Ethiopia’s currency devaluation. What was surprising and interesting about this move is that the devaluation was not undertaken under the usual duress of “macroeconomic adjustment.”

Typically, in Africa, macroeconomic and foreign exchange crises have been the trigger for devaluation. A devaluation helps because it increases exports and reduces imports, thereby increasing the foreign exchange position of a country; and it also reduces domestic spending and brings it more in line with a country’s production.

In this instance, however, the devaluation seems to target structural change, to boost the tradable sector so that it can provide the basis for long run growth. Chris thinks that the devaluation—especially since it was unexpected—might create investor uncertainty. So, he sees it as a trade-off between promoting structural change and engendering a climate of uncertainty.

I see it differently. Tradable sectors and exports can indeed be key for development. And sub-Saharan Africa’s tradable sectors are handicapped by aid and natural resource revenues, which tend to promote non-tradable sectors and encourage consumption over production (see the evidence in my paper with Raghu Rajan on this). Moreover, countries in the past that have grown sustainably have had vibrant exports sectors and have had competitive exchange rates (see the evidence in my paper with Simon Johnson and Jonathan Ostry and also in this paper by Dani Rodrik).

So, three slightly different takes on this Ethiopian move would be the following. First, this devaluation can be seen—not as actively favoring or even subsidizing some sectors as it would be in the case of China, for example—but as offsetting a previous distortion (aid and resource revenues).

Second, instead of viewing this as creating investor uncertainty, it can perhaps be seen as a credible and durable pre-commitment to promoting structural change (provided of course future actions are consistent with this move). The private sector can be assured that there would be durable advantage in investing in the tradable sector.

Finally, the devaluation is heartening if it reflects a realization on the part of African policy-makers that the key to development is structural change but one that is brought about in a market-friendly manner rather than in the dirigiste manner of the past. Watch out for more such moves by other countries.

8 thoughts on “Ethiopia’s experiment (continued, by someone actually knowledgable)

  1. Hardly convincing. I would have thought that a devaluation which favoured some sectors could be described as a good industrial policy as long as the favoured sectors could serve as engines of growth in the economy. Additionally, if aid is what is creating the distortions in the values of the real exchange rates then it is preferable to deal with that directly. Devaluation is certain to lead to higher rates of inflation as is currently happening in Ethiopia. What about people who are thrown into poverty due to devaluation-led price rises? The IMF has always turned a blind eye to the huge social and political costs of devaluation in Africa. Such policies immersed many African states into civil strifes and deepening poverty. It appears that old habits die hard!

  2. I think the mere fact of promoting exports cannot induce structural transformation. At minimum since the main export of the country has been agricultural products, this facilitates specialization in dynamically less productive economic activity. Indeed, this point was noted by Matsuyama (1992). If devaluation, in this context, is to facilitate structural change, it needs to be accompanied by industrial policies aimed at self discovery as well as removal of deliberate policy induced market distortions in the modern sector. As far as I can see, there is no effort in that direction. In my view, such devaluation aims at static gain at the dynamic cost of less diversification. Moreover, given that the policy is driven by the government without the demand from the business sector or in consultation with the private sector, it seems to add more policy uncertainty given the history of the administration in overnight changes in policies. In the end, it doesn’t matter whether we see and phrase it as incentive for structural change. What matters is how the business sector would perceive this change, which as far as I know is seen with suspicion.

  3. I have to agree with these comments above: What specific market actors are driving a devaluation here, as opposed to it just being a ‘good idea’ if you’re a developing country policy maker? Specifically, regarding Subramanian’s third point–wouldn’t any business could perceive this move as being probably followed by some dirigiste maneuver that would favor one industry or the other. If the devaluation was so much higher than analysts thought it should have been, why would I assume that there are future market-oriented policies/reforms forthcoming? I don’t see how random macroeconomic policy shifts that come about with no discernible immediate (that is, if the evidence has been around for a while, why do this now?) impetus and from a government that has a history of acting erratically could be either a hint of stability or an indicator of realization of market forces. That doesn’t mean that investment of some sort wouldn’t follow, but the character of that investment might be called into question; and the handling of that investment would demand quite a bit of agility from the government to translate into broader development, is that forthcoming here? Also, on the pain in the ass front, what African governments are possibly taking notes on Ethiopia’s macroeconomic policies and thinking about how they’re going to follow suit? (I really don’t know if there are any, so if it’s more than just a rhetorical tool, please let me know.)

  4. I couldn’t agree more with you. I’d add that the statement that “countries that grew had vibrant export sectors/competitive exchange rates” is not enough to evaluate whether these are important for growth. We also need to know whether countries that didn’t grow have competitive exchange rates and strong export sectors (so that we can say how much of the difference of growth between countries can be attributed to exchange rates and export sectors; and still, we wouldn’t have completed a causal empirical exercise). Also, on the “who are you growing for” issue, I’d add that a growth model with devaluated exchange rates is a model where you grow with higher prices, and high protection (through exchange rates) of sectors without comparative advantage (probably, in Ethiopia, capital intensive sectors). Those mean relatively higher capital rents and lower real wages (probably translating in a “growth for the rich” model). I’m not sure whether this is a desirable model of growth, both on equity and efficiency grounds.

  5. I just returned from Ethiopia yesterday after conducting research on the emergence of large-scale commercial farms in the country. I have a few thoughts about this topic based on my time there, especially relating to the second take on the devaluation expressed by the guest poster.

    The two big sectors in Ethiopia are agriculture and services (43% and 45%, respectively). The main tradable goods are high-value export crops (oils, coffee, tea, spices, etc.), and these goods are primarily the province of larger commercial farms. In the interviews I conducted, I heard the same story from many commercial farm investors about government officials reiterating to them that their primary purpose, in the government’s eyes, is to bring foreign currency into the country. Regional government officials regularly check on their progress in that arena, and farms that fail to bring in foreign currency are shut down. The government is more strict about this than they are about enforcing environmental laws, worker safety laws, etc.–despite the fact that representatives come to check all of these things. Moreover, documents from the Ministry of Finance and Economic Development stress the significance of commercial farming in the country’s development strategy specifically as a means to attract foreign currency.

    I also have a few thoughts about the degree to which this was an unexpected move.

    There are lots of avenues of communication between commercial farm investors and the government. For starters, the commercial farm investor with the largest total landholding in the country, Sheik Al-Amoudi, is a very close personal friend to Meles Zenawi. More generally, there are regular meetings between the Regional Governments’ Land Authority offices and private investors. These meetings sound (based on the accounts we heard) like high-level shareholder meetings, in which much attention is devoted to the needs of and challenges faced by commercial farm investors. While it’s total speculation on my part, I wouldn’t be surprised if this move was either a) known of ahead of time by the largest investors, or b) a direct response to the concerns expressed by investors. If you look at the list of incentives (dirt-cheap interest, tax holidays, duty-free capital goods imports) the government already advertises to investors, it’s clear how willing they are to take steps to attract this type of investment. It seems to me this could be one more step intended to help those same investors (several of whom reported that they are struggling to find stable export markets for their product).

    I would be curious if Professor Blattman has a different take on any of this from his time there studying a different sector. What I do know is that government documents are firm in asserting that the EPRDF sees commercial agriculture as the engine of growth for the coming decade, and this devaluation strikes me as a logical move to promote that development strategy.

  6. I too was surprised that the devaluation exceeded the black market rates. But anyway, if we manage to get away with just a slight increase in inflation, then it will have been worth it. If on the margin it’s enough to make a significant impact on exports. We’ll see what happens.

    Like another commentator said, the fact that exports are mostly agricultural will of course mute the impact. It should be noted the major bottleneck to export promotion (of any sector) in Ethiopia is high transport costs due to a combination of being landlocked and not having a (working) railroad. I don’t know what incentives will overcome this bottleneck. The land is highly productive yet almost free of charge. Labour is as cheap as one can get, loans aplenty, tax holidays etc. Yet not enough commercial investment. What is this telling us?

  7. It’s easy to understand how devaluation helps exports, encourages import substitution, increases FDI by making labor and other services cheaper for people with foreign currency, and that the side effect is imported goods will be expensive. It’s simple division and multiplication. But what I still do not understand is how the authorities came up with the 20% figure and how Ethiopia, which virtually exports only agricultural products can think of substituting imported goods? Setting up factories which substitute imports would take too long for the side effects to overshadow the benefits, would they not? Can somebody help me on this?