The fear of Dutch Disease from Guyana’s petroleum development

0

One of the possible harms of an oil export boom is that the rise in oil earnings leads to increased public and private spending, which in turn leads to a sharp appreciation of the real exchange rate, and then to a decline in non-­oil exports and to slower economic growth. This pattern is called the “Dutch Disease,” and is named after the overvaluation of the Dutch guilder in the wake of a boom in the Netherlands’ natural gas earnings in the 1960s. The frequent counsel given to oil states is, therefore, to refrain from spending much of the increased oil earnings, and rather build up financial assets, in order to minimise real exchange rate appreciation.

The proposed mechanism is understood in an economic framework that draws the distinction between internationally traded goods on the one hand, and non-traded goods and services on the other. When spending increases following an oil boom, the increased spending falls both on the traded and non-traded goods. Traded goods include sectors such as cash agriculture and manufactures (processed foods), which are traded in regional and world markets. Non-traded goods include food production for local use (rice, sugar, other crops) or local services.

The dollar price of traded goods is set in international markets. The dollar price of non-traded goods, on the other hand, adjusts to clear supply and demand of non-traded goods at home. The increased demand for trade goods can be met through increased imports. The increased demand for non-traded goods, however, must be met by increased local supply. The price of non-traded goods rises relative to the price of traded goods in order to equilibrate increased supply with increased demand. The rise in the relative price of non-traded goods to traded goods (or equivalently, the fall in the relative price of traded goods) is termed a real exchange rate appreciation.

Therefore, for Guyana, the impending oil boom will raise the total output of the traded goods, this will be equalled to the sum of the non-oil traded goods plus the oil production. Hence, as the entire production of the local economy increases by the amount of the oil boom; the real exchange rate appreciation will induce a readjustment of output in the non-oil part of the economy. With a rising relative price of non-traded goods, workers and capital shift into non-traded goods production. Those workers and capital arrive in the non-traded goods sector by leaving the non-oil traded goods sector. In short, the rise in oil spending induces a shift of production away from traded goods (e.g., cash agricultural and manufactured export goods) and toward non-traded goods and services.

These adjustments are not really a “disease” per se. The rise in non-traded production at the expense of (non-oil) traded production does not by itself constitute a “mistake” of market forces, but rather the only way that the economy can enjoy more of both traded and non-traded goods. The increase in traded goods is met through increased imports. The increase in non-traded goods and services can be met only through an increased domestic output of those goods and services.

These resource shifts can become a true “disease” or market failure if there is something special about the traded goods sector that is being squeezed. Suppose, for purposes of illustration, as is known, that the Guyanese economy is exporting fish and timber before the oil is discovered. Once oil is discovered, workers and capital goods are induced to leave the international fishing and timber sectors and to migrate to the non-traded goods sector. If the fishing and timber sector were making a special contribution to growth (e.g., by spreading international best practices in technology and logistics), the decline of the fishing and timber sectors could spell trouble for the economy at large.

The oil boom would, therefore, induce a decline in a technologically leading sector of the economy, with adverse consequences for long-term growth. One solution would be to limit the boom in oil spending and thereby limit the spillover of workers from fishery and forestry to non-traded goods. Another possibility, however, would be to provide specially targeted subsidies for the two sectors, to support the transfer of technologies taking place in those sectors. A squeeze of the non-oil tradable sector might, under some circumstances, also have special adverse consequences for income distribution, particularly hurting the poorest of the poor. That is less likely than often supposed, however, since the poorest of the poor are often economically isolated rather than in tradable goods production.

Moreover, the advice to save rather than invest the oil income in order to protect the poorest of the poor would not make sense in any case if the public investments have direct benefits for the income-earning opportunities of the poorest (e.g., by expanding the road and power grids into impoverished regions).

The real fear of the Dutch Disease, in short, is that the non-oil export sector will be squeezed, thereby squeezing a major source of technological progress in the economy. Nonetheless, this fear is vastly overblown if the oil proceeds are being properly invested as part of a national development strategy. Suppose that the proceeds of the oil earnings are being invested in infrastructure (roads, power, telecoms) that raise the productivity of workers in both the traded and non-traded goods sectors. Assume for the moment that all of the investment goods are directly imported by the government using the oil proceeds. There is no direct spending effect of the oil income. Consumption rises to the extent that the non-oil sectors (both traded and non-traded) expand following the increased public investments financed with the oil income. Production and consumption of both non-oil traded goods and non-traded goods increase. The real exchange rate may or may not appreciate relative to the initial equilibrium level of the economy, but it does not matter very much since the non-oil traded goods sector expands; in any event. It expands as a result of the increased productivity due too public investments (Humphreys, Sachs, Stiglitz 2007).

If the increased spending on public investments falls partly on non-traded goods (rather than entirely on imported capital goods), there can be a minor Dutch Disease effect emanating not from a consumption boom but from the investment boom itself. In this case, the rise of investment spending will tend to lead to an appreciation of the real exchange rate and a short-term squeeze on tradable goods, at least until the productivity-enhancing effects on tradable production kick in. This can be quite fast, however, since the benefits of roads, power, and other infrastructure investment spending can come online very rapidly. Any squeeze on tradable production is likely to be very short-lived.

It is also quite possible, especially in the poorest countries, that the oil boom leads to a real exchange rate depreciation if the public investment financed by the oil substantially raises the productivity of the non-traded sector. This can be a very important and likely outcome. In the poorest countries, staple food production (e.g., maize) is the most important non-traded good in the consumption basket. Although we often think of these staples as internationally traded goods, in fact – because of very high transport costs in the rural areas of impoverished countries – staple food is consumed mainly on the premises of the farm household, rather than being marketed and traded for other goods. Food also constitutes by far the largest single item of household consumption of the poor. If the oil earnings are invested in raising the productivity of smallholder farmers (e.g., by financing improved seed varieties for local production), then the production possibilities of the local economy will increase. The overall effect of the oil export boom may be a reduction of the relative price of non-tradable foodstuffs and therefore a real depreciation. Moreover, it is clear that the production of both non-traded and (non-oil) traded good increase. There is, once again, no squeeze of non-oil traded goods.

In summary, the Dutch Disease is a worry mainly if the oil boom is used to finance consumption rather than investment. In that case, the non-oil traded sector might well be squeezed on a sustained basis, with adverse consequences for long-term growth. This is very unlikely if the oil earnings are properly used for public investments in economies largely bereft of public goods, especially infrastructure. In that case, the positive benefits of increased public investments on the non-oil traded sector are appreciation.

A final note on public investment is warranted here. Even when public infrastructure (roads, ports, power) is highly productive, and when financing is available, the actual physical investments will necessarily take time to put in place, and the optimum pace is itself an economic calculation. Many investment projects impose adjustment costs (e.g., disruptions of other economic activities or congestion due to the investment projects) that increase in proportion to the rate of investment.  The optimum response, in that case, is to spread the investments over time, to maximize the benefits of the investments net of the adjustment costs themselves. This pacing of investments is sometimes described as investing according to the “absorptive capacity” of the economy.

Perhaps the most famous example of an investment boom gone awry was the massive and costly congestion in Nigeria’s ports in the spending boom that followed the oil price increases in the early 1970s. The optimum pacing of investment spending is not motivated by the Dutch Disease per se, or by any automatic desire to spread oil spending over time, but rather by the adjustment costs imposed by the investment projects themselves.