Perceptions of Growth

The concept of economic growth, as the driver and goal of development, is bound up in an intricate labyrinth of conflicting ideas and viewpoints. Alex Cobham unravels some of the complexities that surround the relationship between growth and development, offering a challenging insight into the role that growth could play in the delivery of real, sustainable development goals.

On the face of it, there shouldn’t be much room for disagreement. All other things being equal, a country reaching a higher level of income per capita should have not only lower levels of absolute poverty but also stronger performance in terms of education, health and other important development outcomes. The only way to achieve higher income per capita – excluding deliberate reductions in the number of capita – is through economic growth.

So – growth is good for development. It follows that what we really need to do is to put our effort into working out what the obstacles are to growth, and what policies best promote it.

Or perhaps not. The opening paragraph would gain broad agreement from policymakers, academics and civil society North and South. The second paragraph, however, is a non sequitur. Higher income per capita is associated with better development outcomes, but it is not the only such factor. And of course all other things are not equal. Policies to achieve growth will affect development goals in other ways also.

The renewed interest in growth follows something of a cycle in policy. Structural adjustment and wider conditionality in the 1980s and early 1990s concentrated in reforming the institutions of the state, to the neglect of investment in the social and economic sectors. In the latter part of the 1990s, the focus shifted to trade policy reforms, alongside renewed investment in the social sectors, to the neglect of basic physical infrastructure as well as the productive sectors such as agriculture.

At no stage was growth in the current period given effective priority. The current pressure to achieve the Millennium Development Goals risks continuing and exacerbating this process - depriving countries of the opportunity to focus more of their limited resources on supporting growth, while emphasising short-term goals that may not be sustainable even if achieved.

The renewed interest in growth may then be a valuable corrective measure. To approach an appropriate balance requires recognising the complexity of real development goals, and hence a better understanding of the potential, and of the limits, to what growth can offer.

Complexities of development

Development is complex. We can ignore this and focus on individual areas of the problem, but ultimately to be effective we – NGOs and policymakers alike – need to recognise and reflect it in our analysis. I take the goal to be sustainable human development, as a base for that analysis.

The UNDP Human Development Report in 1990 defined human development as ‘a process of enlarging people’s choices. The most critical ones are to lead a long and healthy life, to be educated, and to enjoy a decent standard of living, with additional choices including political freedom, human rights and self-respect’.

Ranis, Stewart and Samman (2005) build a list of human development categories from this quotation and the considerable literature in the area, which gives a clearer – though necessarily subjective – view of the component factors of human development:

The HDI [the UNDP’s Human Development Index], which includes health, education and a measure of income (i.e., it broadly covers bodily health, literacy and basic aspects of material well-being)
Mental well-being (i.e., an individual’s psychological state)
Empowerment (particularly of the deprived)
Political freedom
Social relations
Community well-being
Inequalities
Work conditions
Leisure conditions
Dimensions of security – political (i.e., freedom from political violence or instability)
Dimensions of security – economic (i.e., freedom from economic fluctuations)
Environmental conditions.

“Sustainable human development” entails the additional qualification that benefits for current generations do not diminish those available to future generations. This qualification entails a range of constraints on policy, from fiscal prudence to, critically, limiting climate change to levels consistent with future development.

The limits of growth

If we simply prioritise growth ahead of this more complex but real goal, development loses out in two important ways. The limits of an unnecessarily growth-focused approach are powerful, and damaging. Growth will be used inefficiently – and its contribution limited.

Policy has often been aimed at maximizing growth itself, instead of the benefits of growth. This leads, in particular, to policy prescriptions that prioritise growth more or less regardless of the impacts on key human development components, for example, inequality, social justice or environmental protection. Inevitably, this gives rise to inefficiently low gains in poverty reduction and/or broader human development.

Imagine a business which becomes confused in a similar way, perhaps recognising the potential contribution to profit of R&D but then becoming confused and unable to distinguish between instruments and goals. Rather than maximising profit, it maximises R&D – and in doing so inevitably arrives at an equilibrium with sub-optimally high R&D and sub-optimally low profit.

The added complexity of the growth-development confusion is that policymakers are unable simply to choose the level of growth they desire. As a result, while the firm in the example will at least end up with high R&D (albeit with low profits), policymakers mistaking growth for a goal in its own right cannot even be sure of delivering this – but will certainly deliver lower returns and hence sub-optimally poor human development.

An obvious area in which this is seen is poverty reduction, where the growth obtained in low-income countries has been markedly anti-poor compared to that in high-income OECD countries.

To see the limits of growth more clearly, consider a particular application. It is often claimed that growth and improvements in the Human Development Index (HDI) are highly correlated, and hence that targeting the former will deliver gains in the latter. Aside from the latter’s limitations as a measure of broad human development, this approach is flawed in any case. The relatively strong correlation that does exist is between levels of income per capita and of the HDI.

Figure 1: “Growth is associated with better human development”

Source: DFID presentation, BOND AGM, 24 October 2007.

Figure 1 shows a scatter plot of per capita GDP levels (i.e. income and not growth) against the HDI (after removal of income per capita, i.e. an index based on health and education outcomes only).

Allowing for the fact that the graph shows an association of income levels rather than growth, we can assess what its implications actually are for the potential contribution of growth. Taking Tanzania as an example, were its GDP to grow by 5% annually, on a completely stable basis, then using (and extending forward) DFID’s forecast of 3% population growth, it would take 37 years for per capita income to double (from $580). This would mean a shift from roughly one twentieth of the first separator on the horizontal axis to roughly one tenth, or from the first red line to the second.

Even assuming such an unrealistically long and stable period of growth, for a period approximating the average life expectancy in a number of sub-Saharan African countries, the implied impact of this growth on HDI in Tanzania would be small. It should be clear that growth cannot be relied upon, even under such extremely favourable assumptions, to provide on its own the improvements in development that we seek.

What if, as suggested above, we decide to target an alternative single instrument? Instead of growth or income, figure 2 shows the association between child mortality rates and the HDI. While the logarithmic relation in figure 1 “explains” around 66% of the variation in HDI, a much higher share (88%) is similarly “explained” by the rate of child mortality. It seems obvious that if a strategy were pursued of targeting a single factor in order to raise development, it would fare rather better if the single factor in question were not growth but rather infant mortality.

Figure 2: Lower child mortality really is associated with better human development

Such a single factor approach would of course be foolish however, even with this more appropriate choice of factor. It clearly makes no sense for policymakers to use all but one of the many instruments available with the aim of maximising the remaining one, instead of focusing their resources on the ultimate goal.

The potential of growth

And yet the potential, and importance of growth should not be underestimated.

The new DFID, under Prime Minister Gordon Brown, has re-emphasised the importance of growth. Speech after speech includes the claim that 80% or more of the poverty reduction since 1980 has been the result of growth. Leaving aside serious concerns about the econometric result underlying that particular claim, it is generally agreed that on average, over the long run (perhaps forty or fifty years), growth in average income per capita has been shared by the poorest and richest quintile in countries in proportion to their original income share. Given growing populations, growth is essential to raising incomes.

Paul Collier¹ highlights another potential contribution. The ‘bottom billion’ not only face deep and persistent income poverty, but also a much higher risk of conflict than most of us. By implication, growth may offer a way out of the ‘conflict trap’ that he identifies. Clearly the determinants of conflict are more complex than this suggests, as the work of Frances Stewart on inequality between groups confirms, but the wider instrumental value of growth should also be a consideration in our analysis.

Often overlooked are the linkages in the other direction – those aspects of development whose weakness must be addressed before growth can take off and be sustained. At the general level, Stewart and Gustav Ranis have shown that countries achieving human development increases before growth in income are much more likely to see sustained improvements in both. Countries achieving only growth first show a systematic tendency to see growth fall off, and without ever achieving the human development benefits. At a more specific level, the World Bank identified income inequality as a key obstacle to growth (in Latin America above all).

Getting growth right means doing the analysis on two fronts: of its interrelationships with various aspects of the real development goals; and of the way to maximise its benefits to these goals, rather than maximising growth itself.

Looking to the future

I took up a post managing the policy team at Christian Aid in October last year. One of my first outings was to the BOND AGM, to outline some views on growth and sustainable development. In contrast to the DFID presentation, I insisted that economic growth be treated as one of many instruments, and not the goal – or even one of the goals – of development.

In a speech on growth earlier this year, Douglas Alexander spent some time explaining that growth is an instrument and not a goal. Whether that counts as an advocacy win, or is just happy coincidence, remains unknown. But does it mean that we all see growth in the same way now?

Almost certainly not – yet. The terms of reference for DFID’s proposed new Growth Centre, for example, frame the questions only in terms of achieving higher growth – with no thought to the possibility of more or less beneficial growth, although their own new policy paper highlights this.

Douglas Alexander has complained, not unreasonably, that growth is a dirty word for some NGOs. It should not be. That doesn’t mean that we should uncritically accept a foolish over-emphasis on this particular instrument; but it is important that we recognise the potential contribution of growth if we get it right – within a broad framework of appropriate polices, and not as a stand-alone or dominant driver. NGOs need to engage to help make that a reality.

Alex Cobham is Policy Manager at Christian Aid.

1. Paul Collier, The Bottom Billion: Why the Poorest Countries Are Failing and What Can Be Done About it (Oxford University Press)