Industrialisation – No One Road Takes us There [analysis] (

I do not know whether it is the right time to write about industrialisation or not. For one, the whole hype has been about the official visit of United States President Barack Obama and a global discussion on financing for development. In a country where collective thinking is a way of life, it often is difficult to dissociate oneself from the popular trend and thinker over something substantial.

Yet, I think deliberating on industrialisation is important, as no series of hypes could actually bring about structural change in an economy. The marginal economic value of hype, is, after all, insignificant.

In my previous Viewpoint about industrialisation titled, “Which Way Should Ethiopia Industrialise” (Volume 16, Number 791, June 28, 2015), I suggested that it would be better for Ethiopia to follow a thoughtful mix of FDI-driven industrialisation and local clustering. Aside from the prevalence and structure of factor endowments, I mentioned key policy issues, such as local ownership, initial advantage and policy convenience, as drivers to a blended approach.

After my viewpoint was published, though, I received comments regarding the assumptions that I had made in the piece. A full-fledged response to my piece headlined, “In Capitalisation Sits the Key to Industrialisation” (Volume 16, Number 792, July 05, 2015), written by Abdulmenan M. Hamza, was also published in this newspaper. In addition, research by one of the high-level politicians of the country, Arkebe Oqubay (PhD) was published by Oxford University Press in a book titled, “Made in Africa: Industrial Policy in Ethiopia”.

Hence, I tried to test my ground against this varied evidence and wanted to elaborate further. Much as I would try to take into consideration the book by Arkebe, I prefer to write my full-fledged review of the book and share it here. But I will make my points with reference to key points he raises in his book.

Often, in discussions about industrialisation, there is a tendency to fall prey to the traditional state versus market debate. This has been one area mentioned in the comments I received, in Abdulmenan’s piece and Arkebe’s book. I would not dare to engage myself in the rather naive endeavour of choosing one or the other. It takes nothing but knowledge in elementary economics to understand that both the market and the state have their own role in the economy, regardless of their level of development. As much as both have limits, both have peculiar leverages. Hence, choosing one over the other is a futile task.

The debatable issue, as far as I can see is, where the unique leverages ought to be deployed. Both Arkebe’s and Abdulemenan seem to advocate that an activist state is a sine qua non for Ethiopian industrialisation. Without it, they seem to think, the markets are not, at least as they stand, able to unleash enough capital to drive industrialisation.

For Arkebe, it is not every active state that could succeed in the endeavour to realise industrialisation, but rather a learning activist state. The level of learning the state engages in, according to him, is guided by the latitude of performance. This means that higher returns to learning, incentivises the state to be more active in policy making.

What is missing in the whole debate, are the limits to the state. It is more than obvious, and is economically well-evidenced, that the state has limits of engagement. It cannot have all the resources to push industrialisation. A limitation in resources means that there are inherent limits to the support the state could offer even under a most protected environment.

Beyond resources, though, the state has limits in terms of incentives and capabilities. Optimal allocation of resources in an economy could not be achieved through a statist approach, for the state lacks the essential incentives to do so. Since the state is driven by political interests, its allocation of resources also follows the lines of interest. This, certainly, would be far away from optimal.

In view of industrialisation, this means that the factors of production will be allocated following the lines of interest that the state represents or stands for, and hence the whole matrix would be far from an optimal combination of resources. By the way, for me, this is the primary reason that the infant industry approach remains a costly path to follow for a country like Ethiopia.

In terms of capabilities, too, the state lacks essential traits, structures and interests that would enable it to leverage all the opportunities that could facilitate industrialisation. In a purely economic sense, this entails lacking structures, such as pricing mechanisms, that the market could leverage to influence aggregate demand and supply. In this sense, the state has only its own demand or supply capability at hand. This means, it can live for the market, but cannot dictate it.

These limitations happen, regardless of the learning curve of the state. No type of state could have these leverages. Essentially, then, the state could not be a force to rely on for industrialisation. It still remains true that the state can (and needs to) play its regulatory role.

If one analyses the Ethiopian economic scene in light of industrialisation, it would be easy to see that fragmented markets are the major roadblocks. One needs no more evidence than the effort it would take a trader to collect a quintal of butter (or honey), a carton of eggs, a sack of lemon or a quintal of spices from a given rural locality. It takes days, if not weeks, to collect pieces of the items to actually get a supply of a certain standard. Neither is there enough aggregate supply – a result of inadequate aggregate demand – nor are there adequate suppliers to fulfil such kinds of demand.

The task is even tougher for a trader that needs any combination of these products. It demands doubling the effort albeit at a cost double the first one.

Ethiopian markets are not just fragmented, but they are also very hierarchical. Such hierarchical markets mean policy guidance could not be effective for the mere fact that it would lose its momentum while passing through the hierarchies.

As illustrated in Arkebe’s insightful book, this fragmentation of market is a crucial roadblock for agro-processing industries and light industries including, leather and textile industries. Not only does the fragmentation affect the stable supply on inputs, ranging from skins and hides to cotton, they are also effectively holding up aggregate demand not to fully express itself. Hence, large industries often get involved in importing inputs they could have accessed locally at low prices and limit themselves to the export market.

Without consolidated markets, it would be difficult to have optimum forward and backward linkages. As Alberto O. Hirshman (widely quoted in Arkebe’s book), considered the father of endogenous growth models, argues, linkages are crucial for an industrial process to translate into a wholesome economic push and bring about structural influence over the economy. Therefore, it is this very important element of industrialisation and structural transformation that is hindered by the lack of consolidated markets.

The issue then, is how far could an active state help establish such a market. Investment by the state in infrastructure could partly help consolidate the market. It can also help bring some level of integration within it. Nonetheless, as much as fragmentation is not a result of infrastructural deficit only, it would not be solved instantly by infrastructure investment.

Essential elements, such as transaction costs, incentives, and price differentials define market integration. There is little that the state could do in direct investment in these areas. And if it does, the distortionary impact would be more damaging.

But in view of the interests of an industrialist, these are essential elements to compete and remain profitable. No smooth infrastructure would compensate for the cost imposed on an industrialist by price differentials. Only a properly functioning market could give her the rationale to live with it.

This does not mean the market does not have limitations. Because of the asymmetry of information embedded within it, the market could also fail to function smoothly. It could also act parochially along the lines of interests. This is where the regulatory role of the state is highly demanded.

What may not have stood out in my previous viewpoint, as could be understood from Abdulmenan’s response, is the fact that my preference for a blended-approach stems from my observation of the limitation of both economic agents. Choosing either FDI-driven industrialisation, which demands integrated markets, or clustering, which happens only under active state support, entails risking the limitation of the determining agent to set the trend of industrialisation. Without even considering the level of development of Ethiopian markets and state, this path is economically risky and unviable.

Challenges of the Ethiopian state, such as excessive opportunism, corruption, undeveloped politics-technocratic nexus, regulatory differentials, lack of meritocracy and so on, mean that relying on an FDI-driven model would be hugely risky. This, of course, is not to mention the fact that it would distort the wealth creation aspect of the economy away from the local people.

Therefore, if Ethiopia has to industrialise, the best way for it is to leverage the role of the state and the market as optimally as possible. It also needs to attend the limitations of both the state and the market as effectively as possible.

As Abdulmenan has shown in his Viewpoint, this is easier said than done. It, however, is not impossible. If countries, such as China, Vietnam and Mauritius, have achieved it, there is no way that Ethiopia cannot.

The stake for Ethiopian policymakers, at least as I see it, is to strike a good balance between policy alternatives. It is not time to live for one alternative or the other as each involves its own risk that the nation does not afford at this time.