Decoding The Bangladesh Paradox — A Research Agenda
The macroeconomic fact is, in the last decade, under all three governments, per capita GDP have grown by around 4½ per cent a year. At that rate, average real (that is, inflation-adjusted) income doubles in 16 years. …. This is impressive stuff, for which every recent government deserves some credit.
That’s the conclusion from the post on real GDP per capita growth under different governments. Of course, real GDP per capita is a means to the end, not the end in itself. What we really care more about is the standard of living that higher real GDP per capita entails —that is, it’s the development record, and not just the growth, under different governments that we want to know.
This, however, raises two questions. First, how do we attribute to any particular government the growth and development record when policies under any particular government are likely to have long term consequences? And second, how do we explain the Bangladesh Paradox:
The belief that growth brings development with it—the “Washington consensus”—is often criticised on the basis that some countries have had good growth but little poverty reduction. Bangladesh embodies the inverse of that: it has had disproportionate poverty reduction for its amount of growth.
That quote is from a November 2012 Economist article. That article, and accompanying editorial, had a go at explaining the paradox. Joseph Allchin had a crack more recently at the NY Times. The suspects are usual: garments, remittance, NGOs. But we economists are a parsimonious lot, or so we like to think. We would like to know exactly what contribution each of these factors made, what was the channel through which the factors affected growth and development, what role, if any, did government policy play, and what all that means for future.
I haven’t seen a comprehensive analysis of the Bangladesh Paradox. And no, I am not going to provide the answer in this post. Rather, over the fold is a research agenda on how to analyse the Paradox.
Let’s recap the Paradox with this table from the Economist:
Source : The Economist
The first thing to explore is whether Bangladesh is compared with an appropriate benchmark? Is it that Bangladesh has done better with its growth and income, or is it India (or Pakistan) that is the exception?
The Economist notes that Bangladesh has a few features that India or Pakistan lacks:
Because of its poverty, it has long been a recipient of vast amounts of aid. With around 150m people crammed into a silted delta frequently swept by cyclones and devastating floods, it is the most densely populated country on Earth outside city states. Hardly any part is isolated by distance, tradition or ethnicity, making it easier for anti-poverty programmes to reach everyone. Unusually, it has a culture that is distinct from its religion: although most Bangladeshis are Muslims, their culture and language are shared with the non-Muslim Indian state of West Bengal. Religious opposition to social change has been mild.
Has Bangladesh received more aid per capita than other poor countries? How does Bangladesh’s growth-development trajectory compare with other densely populated monsoon deltas — say, countries along the Mekong? Or perhaps, Bangladesh should not be compared with India and Pakistan as a whole, but with the four Pakistani provinces and 32 Indian states?
Does a paradox remain if the comparators change? Does Bangladesh still perform better in terms of development / living standard given its growth / level of income? A proper research agenda would answer these questions first.
Suppose the answer is yes, that the Paradox still remains, its resolution will rest on a two-part investigation.
The first part would explore the GDP story in detail. That Bangladesh does better given its GDP does not make GDP irrelevant. Quite the contrary. Bangladesh was wretchedly poor place until the 1980s. It’s not a coincidence that things started getting better as the economy started accelerating. We would want to know what about the GDP growth process that may have contributed to the development in a relatively favourable manner.
What do we know about the growth story of the past few decades?
From a strictly growth accounting perspective, we know that while favourable demographic transition and female workforce participation have helped, it is multifactor productivity that explains the GDP acceleration. From a sectoral perspective, we know that agriculture’s share of the economy has shrunk, that of manufacturing has risen, and services have become more productive. And we know the particular industry that has led the charge — readymade garments.
So far, this seems like a straightforward export-led manufacturing driven growth story a la our neighbours to the east and north.
Yet, it’s not so clear cut when we look at the expenditure side of GDP. Unlike the Asian fast industrialisers, Bangladesh has not experienced an investment boom. In fact, low private investment relative to GDP may be the single most important problem facing the country’s economy.
Of course, why investment hasn’t grown is a question that needs further exploration — and the answers will have obvious policy implications. But is there something to the consumption pattern as well? Particularly, have remittance and microcredit affected consumption above and beyond what would be implied by wages growth coming from industrialisation?
In addition to the macro trend, do industrialisation, remittance and microcredit interact in a way that have microeconomic — that is, household and firm level — impact favouring consumption over investment even after accounting for various market and government failures that inhibit investment?
What am I getting at here?
Well, how about a stylised, and very speculative, story along this line — while RMG has meant women entering the formal workforce, migrant worker boom has sent a lot of risk-taking men overseas; aided by the NGOs and microcredit, households have smoothed consumption and invested in human capital of their children; but they have not invested in physical capital, avoided entrepreneurial activities, and have not pushed for a more investment-friendly polity.
We would want to explore this story further. We would also want to explore the income side of GDP, and tie it into a political economy analysis.
The remittance boom, for example, should see the labour share of the economy rise. Of course, the question is, what happens to the money that is remitted back? It’s reasonable to assume that unskilled labourers are from the poorer parts of the society. So, in the first instance, any remittance back to the villages is a good thing in that it reduces the direst type of poverty — that is it stops things like famine or malnutrition. But what happens after that? My tentative hunch is that a lot of remittance has been saved but not invested in a productive way, rather they ended up fuelling land/stock prices —this is an area that needs to be explored in detail.
What about the RMG boom?
Theoretically, proceeds of the manufacturing boom should accrue to both labour and capital. Has that happened? Has the process of distribution been dynamic or static? Here, by dynamic I mean whether the industries are going up the value chain —from the cheapest tee-shirts to more expensive designer brands to leather and other fancier fashion items to toys to cheap electronics to expensive electronics to stuff that requires more skilled labour. If the process has been dynamic, then we should expect less tension between labour and capital, because both wages and profits rise over time.
In addition to the detailed exploration of the growth process, the research agenda will need to focus on the factors that explain the development above and beyond what might be expected from the growth itself.
As a starting point, let’s take the four factors listed by the Economist: government spending and policies on social programmes that assisted family planning and empowered women; the green revolution; remittance; and NGOs.
Let’s think about these factors in a systematic way.
Source: The Economist
The chart on the left from the Economist illustrates the female empowering social transformation. But how important has the government, and NGO, interventions been relative to the advent of the RMG sector. As far as girls’ education is concerned, Mushfiq Mobarak of Yale finds the garments made the difference (this is a subject of a detailed post). We would like to see the relative impacts of industrialisation, direct government policies, and NGO activities analysed across various metrics.
The green revolution is a relatively straight forward story, as is its impact, and remittance we have discussed above. In addition, we would want to know how, if at all, the impact of these factors have changed (and is likely to change) in a more rapidly urbanising Bangladesh.
Finally, we would want to analyse the economics and political economy of the NGOs —what the Economist calls the ‘magic ingredient’.
Large NGOs such as BRAC are as much business conglomerates as philanthropies. In fact, the Economist compared BRAC to Korean chaebols. Is that a reasonable comparison? Do we understand the microeconomics of NGOs? Does our view of Bangladesh Paradox change at all if we viewed these NGOs as little different from Korean or Japanese business houses at comparable stages of development? And what about the impact of the NGOs on public finance, and indeed the state’s capacity to build institutions in general?
Needless to say, this is a pretty ambitious research agenda. But it’s hardly impossible. Is there anyone out there to tackle this?
(First published at AoD).