Johan Fourie's blog

I'd rather be a comma than a fullstop

Archive for February 2017

Will a new John Maynard Keynes please stand up?

with one comment

keynes

In his 1936 book that would give theoretical support to Franklin D. Roosevelt’s New Deal, the British economist John Maynard Keynes wrote the following: “The ideas of economists and political philosophers, both when they are right and when they are wrong, are more powerful than is commonly understood. Indeed the world is ruled by little else.”

There is little doubt that the ideas of economists – their models or theories or ideologies – have had a profound influence on our daily lives. To give just one example: when South Africa’s Monetary Policy Committee decides whether to lower or raise the interest rate, they do so based on a model that predicts the likely effects of the policy change.

But there is not one universal model. Macroeconomists, depending on their school of thought, have different beliefs about how the real world works. Some, like those that propound Real Business Cycle (RBC) theory, argue that business cycle fluctuations are the result of exogenous changes to an economy, and because wages and prices are flexible, there is little that governments can do through fiscal and monetary policy in a downswing. Others, like New Keynesians, see a greater role for government intervention because of things like sticky wages and prices.

Though these ideological disagreements in macroeconomics are nothing new, one would expect that the competition between these ideas would ultimately lead to a better way to explain what is happening in the world. This is, however, not the case, and one of the main reasons the World Bank’s new chief economist, Paul Romer, formerly professor at Stanford and New York University, recently wrote a stinging critique of macroeconomics. “I have observed more than three decades of intellectual regress”, he begins, explaining that macroeconomics has gone down a wrong path and have now reached a dead-end.

The basic premise is this: macroeconomists’ increasingly sophisticated models have ignored real-world evidence. Simon Wren-Lewis of Oxford University explains: “If we look at the rise of Real Business Cycle (RBC) research a few decades ago, that was only made possible because economists chose to ignore evidence about the nature of unemployment in recessions.”

Why would scholars willingly choose to ignore real-world evidence? “The obvious explanation is ideological” says Wren-Lewis. “I cannot prove it was ideological, but it is difficult to understand why – in an area which … suffers from a lack of data – you would choose to develop theories that ignore some of the evidence you have.” Ironically, this shift came at a time when the rest of the profession, fields like labour economics, development economics and international trade, shifted the other way: from theory to empirics.

To be fair, RBC models have become less popular, replaced by ‘New Keynesian’ models. These ones make more realistic assumptions of the real world, like adding sticky wages, but still share many of the same concerns. Because these new generation models, known as Dynamic Stochastic General Equilibrium models, are ‘built from the ground up’ (in other words, built on microeconomic foundations) they are far more appealing theoretically (and technically more difficult to construct) than the structural models of the previous generation, models which only considered the interlinkages of different macroeconomic variables. But in practice they turned out to be less than satisfactory: DSGE models require the modeller to make many assumptions about how individual agents behave, and it turns out that this behaviour is difficult to measure, making the assumptions rather subjective. The result is that, though theoretically plausible, the outcome is often far removed from the real-world evidence. And, most importantly, forecasts – that thing that most people think is an economist’s only job – are often far less accurate with DSGE models than with multiple-equation structural models. That is why most reserve banks, including our own, still mostly rely on structural models to help predict the future outcomes of their policy decisions.

So why continue with DSGE models? This is a consequence of the way scholarship works. In a blog post in response to Romer, Narayana Kocherlakota of the University of Rochester writes: “We tend to view research as being the process of posing a question and delivering a pretty precise answer to that question. In this process, machines that can be used by many scholars to generate answers to wide ranges of questions are highly prized. The King-Plosser-Rebelo real business cycle model was one such machine.  The New Keynesian three-equation model is another. And people who have the mathematical and computational skills to make machines even more powerful, so that they can answer even more questions, are naturally highly valued.”

Yet if these ‘answers’ increasingly abstract away from reality, we are not much better off than before. Keynes, who developed a new theory when the existing paradigm could not explain the real world anymore, knew this all too well: “Practical men, who believe themselves to be quite exempt from any intellectual influence, are usually the slaves of some defunct economist.”

For academic macroeconomics to regain its influence, it needs to be, as Kocherlakota writes, “a lot more flexible in our thinking about models and theory, so that they can be firmly grounded in an improved empirical understanding”. This won’t be easy. As Locherlakota says, it will be “hugely messy work”. Will this generation’s John Maynard Keynes please stand up?

*An edited version of this first appeared in Finweek magazine of 9 February.

Written by Johan Fourie

February 26, 2017 at 18:24

Four high-growth scenarios for Africa

leave a comment »

africagrowth

Can African countries sustain the relatively high growth rates they attained since 2000? At the start of 2017, putting aside the newsworthy political shifts and the fear of many that the developing world has entered a ‘secular stagnation’, this remains the most vexing question for those of us on the African continent.

It is not a question with an easy answer. The stellar economic performance of several African countries has created an ‘Africa rising’ narrative where further progress – and catch-up to the developed world – seems inevitable. A more pessimistic counternarrative argues that this growth, from a low base, is largely the result of favourable commodity prices and Chinese investment. Both narratives had, unfortunately, made little use of either economic theory or history.

Enter Dani Rodrik, professor of International Political Economy at the John F. Kennedy School of Government at Harvard University, who tackles this question in a new paper in the Journal of African Economies. He first shows that many African economies have indeed improved since 2000, but that many, including Senegal, the DRC, the Ivory Coast and Zambia, remain on levels below those immediately following colonialism (around 1960). The second fact he establishes is that the rapid growth of the last dozen years has not lead to a large structural transformation of the economy. Whereas rapid growth in south-east Asian economies during the late twentieth century resulted in the growth of manufacturing, a more productive activity than subsistence farming, high growth rates in Africa have not had any effect on the relative size of manufacturing. In fact, in many countries, the size of the manufacturing sector has actually declined since 1975.

Rodrik attributes these changes not so much to factors unique to Africa – like a poor business climate or weak institutions or bad geography – but to a global trend of deindustrialisation. Even Vietnam, a country which has recently experienced rapid growth, has not seen much growth in manufacturing.  And Latin American countries, which have decidedly better institutions than three decades ago, have also not seen much growth in manufacturing. Technological change – the move to automation, for example – is one likely reason.

So despite high growth rates, African countries have not industrialised – and, in fact, may have even begun to deindustrialise. This is why Rodrik is pessimistic about Africa’s future growth prospects. He nevertheless concludes by considering potential scenarios in which Africa can indeed sustain high growth, and identifies four possibilities: 1) To revive manufacturing and industrialise, 2) To generate agricultural-led growth, 3) To generate service-led growth and 4) To generate natural resource-led growth.

Let’s start with agriculture. Although many African countries have a lot of potential to expand their agricultural sectors, productivity in the agricultural sector remains low. Many farmers are subsistence producers, with low economies of scale. Such a scenario will require a reversal in the current trend away from agriculture. A recent study by Diao, Harttgen and McMillan show clearly how the share of agriculture is falling, particularly as women older than 25 are moving to the cities and into manufacturing and services. This trend seems irreversible, even if changes to technology (like seed varieties or market access opportunities) or institutions (like private property) are made, which means that an agricultural-led high growth scenario seems highly unlikely.

A natural resource-led strategy also seems unlikely for most African countries. Yes, most countries on the continent are well-endowed with resources, but the problems of the Natural Resource Curse and Dutch Disease are well known. It may be an option for some small economies, like Botswana has shown, but one has to question to what extent it can be sustainable beyond a certain level of income.

A third option is to reverse the trend of deindustrialisation. Because a growing manufacturing base seems to be, at least if we consider past examples of industrialisation, the only way to increase labour productivity over a sustained period of time, this is the option preferred by many development agencies. Yet there are many obstacles in the way of a thriving manufacturing sector, including poor infrastructure (transport and power in particular), red tape and corruption, low levels of human capital, and political and legal risk. But as explained earlier, Rodrik believes that even if these (very difficult) barriers can be overcome, it is not clear that manufacturing will return. The Fourth Industrial Revolution may completely alter the nature of manufacturing away from absorbing unskilled labour to capital and knowledge-intensive production. As I’ve said before, it is dangerous to follow a twentieth-century blueprint when production technologies are so different.

That leaves us with one scenario: services-led growth. Services have traditionally not acted as an ‘escalator sector’ as Rodrik explains. The problem is that services typically require high-skilled labourers, one thing that is in short supply in a developing economy. Rodrik does acknowledge, though, that the past will not necessarily look like the future. “Perhaps Africa will be the breeding ground of new technologies that will revolutionise services for broad masses, and do so in a way that creates high-wage jobs for all. Perhaps; but it is too early to be confident about the likelihood of this scenario.”

I don’t see an alternative, though. Yes, some countries, like Mozambique or Tanzania, will be able to expand their agricultural sectors – but higher productivity will probably mean larger farms with fewer workers. A few small countries will be able to benefit from natural resources – from diamonds to rare minerals like tantalum (used in cellphones and laptops); oil-producing countries will struggle, though, as the cost of renewable energies keeps falling. And some coastal countries may even develop their manufacturing sectors, like Ethiopia and South Africa. But for most of Africa, services offer the only reprieve from low productivity, low-wage jobs. From semi-skilled jobs like call-centres and virtual au pairs (apparently the next big thing) to professional services like accountants and designers and programmers, digital technologies must help leapfrog the barriers of poor infrastructure, bad geography and weak institutions. If it cannot, Dani Rodrik’s pessimistic vision of Africa’s future is likely to come true.

*An edited version of this first appeared in Finweek magazine of 26 January.