The economic case for natural resource funds is surprisingly weak. There is, however, an important political rationale for resource funds. The importance of natural resource funds (NRFs) lies in the effects that they have on the incentives facing political actors. If, as is sometimes the case, NRFs do not substantially affect political incentives then they can be ignored or bypassed by governments and have no beneficial effect. To be effective, it can be argued that:
- Withdrawal decisions should be regulated in part by clear rules rather than general guidelines,
- Key decisions should be made by bodies representing the interests of diverse political constituencies, and
- There should be high levels of transparency regarding their status and operation; in particular, there should be a unified budgetary process and public reporting of payments, holdings, and investments.
However, it must be emphasised that in all cases the impact of the institutional details will depend on the extent to which they alter the incentives facing political actors. Despite popular belief in their utility, research on natural resource funds does not find evidence that funds lead to better management of natural resources (Davis et al. 2003; Fasano 2000). Natural Resource Funds (NRFs) are employed in countries that use revenues well, but they are also employed in countries that use them badly. While they may restrain government over expenditure in some cases, in other cases, fund rules are changed or the NRFs themselves are raided when governments wish to increase expenditure. In cases where there appears to be a positive relationship between the presence of NRFs and expenditure smoothing, it is not clear that the NRFs themselves, rather than other features of a country’s political system, are responsible for the policy choices. Norway, for example, is often cited as a country with a highly effective NRF yet inferring a causal effect for the Norwegian case is difficult: the restrictions imposed upon policymakers by the Norwegian fund are extremely weak. In contrast, much effort could be put into ensuring that an economy in transition such as Guyana, have in place a fund for future generations, as a precondition for certain level of international financing of developmental projects. However, governments may feel free to simply change the rules whenever they want greater access to the oil revenues. These facts then create a puzzle: what features, if any, of a natural resource fund might lead to beneficial changes in economic policy choices?
One reason for the poor empirical relationship between NRFs and effective policy is that the economic logic used to justify the use of NRFs does not, by itself, show that NRFs are necessary. Good expenditure policies can be adopted with or without formal NRFs. More important than the economic rationale are political economy considerations. To address the question of whether resource funds are useful, then it would be better to analyse how political incentives influence key decisions regarding how much to accumulate. Setting up a natural resource fund does not automatically change the political economy incentives for “misbehaviour.” The potential value of an NRF lies instead in the details of a fund’s institutional procedures and on how these affect the political incentives facing policymakers. An NRF is useful only insofar as it improves these incentives; if incentives remain unchanged the NRF will not contribute to better fiscal management and might even make matters worse by adding complexity and reducing transparency.
The crux of the argument is the following. A crucial reason for the inability of policymakers to save windfall revenues for the future is that they risk losing control over how the money will be spent at a later time. If a new government can come to power and dramatically alter the way the money is spent, an incumbent policymaker has an incentive to spend more money now, even if he himself would also prefer to smooth spending out over future time periods. This means that some compromises that benefit all political factions are in principle possible although in practice they do not get made. These compromises would require potential future governments to desist from changing policy too much, and this would induce incumbent governments to save more for the future. The problem is in getting policymakers co commit now to implementing more moderate policies in the future. Such commitments are made possible by political institutions. If Norway manages to save most of its oil revenues, it is because the general institutional environment (and not just the oil fund) endows the politics with a high degree of predictability. In countries with weaker governing systems, the challenge is to build institutional mechanisms for commitment and predictability. This point is certainly much more general than the question of whether or not to have an NRF; it goes to the heart of state-building. The more modest focus is on whether and how the occasion of setting up an NRF can be used to make incremental improvements to the institutional structure of a country’s political economy.
In doing so, emphasis must be placed on the risk that the goals on which an NRF is set up to achieve may come to nothing if the political incentives work against them (Humphreys and Sandhu 2007). Therefore, it is crucial for designers of natural resource funds to take into account the political economy context within which the resource fund is to be set up.
Economic Case for Accumulation and Natural Resource Funds
The economic arguments usually advanced for natural resource funds (NRFs) do not, in fact, provide a rationale for new government institutions. Instead, they provide an argument for the principle of expenditure smoothing. The argument runs as follows. In low- or middle-income countries (and a small number of high-income countries) whose main exports consist of fuel or mineral resources, governments often face revenues that are both large (as a share of the country’s economy or compared to the government’s other sources of public revenues) and extremely unstable.
The instability of natural resource rents derives in the short run from the high volatility of world commodity prices, and in the longer run from the fact that natural resources are depletable and therefore cannot be exploited in perpetuity. These features create a special problem for fiscal policy, as volatility in expenditures is generally suboptimal (Davis et al. 2003). Public spending yields “diminishing marginal benefit” – the social gain from spending more than the long-term average in some years is not great enough to outweigh the social cost of having to reduce spending below the average in other years. Such boom-bust patterns, however, are a nearly universal experience in commodity-dependent economies.
This has several important implications. One is that spending should be stabilised and should not track revenues closely. Another is that temporarily large revenues-such as those deriving from depletable natural resources – ought to be saved so as to also benefit future generations. Natural resource funds are often set up with one or the other of these as the main goal, and the detailed accumulation and spending rules designed accordingly. Indeed, NRFs are often labelled “stabilisation funds,” “savings funds,” or “future generations funds.” The general principle behind both stabilisation and saving is the same. When revenues are front-loaded, as is usually the case with natural resource revenue streams, a policy of constant public expenditures does double duty, as it fulfils both a savings function and a stabilisation function.
The main point to note is that the optimal pattern of public spending is independent of the shape of the revenue stream. The implication is that the public administration in such situations must manage the mismatch between incomes and expenses: It must, so to speak, save in plentiful years to compensate for meagre years. The difficulty of this challenge is illustrated by the dramatic failures of most natural resource-rich countries once the meagre years set in.
The need to separate the pattern of spending from the pattern of income means that good fiscal policy in countries with large natural resource wealth typically involves accumulating large amounts of revenues for future use. In a notional sense, therefore, natural resource-rich countries should always have a “fund,” meaning simply that they should have a stock of accumulated savings, to be drawn down when the natural resource revenues dry up. Nothing in this economics argument, however, requires that the natural resource monies be administratively separated in any way from other government assets, as a formal NRF does. In other words, there is no economic need for new institutions or rules governing the accumulation of revenues – there is just a need for accumulation according to some optimal policy. Nor is there any obvious economic reason not to have a fund. Once the optimal policy is determined, it is relatively straightforward to design a fund with rules that in principle carry it out (Engel and Valdes 2000). The relevant criteria for whether to have an NRF or not are therefore those of political economy, not of only economics.
Nevertheless, several countries have sought to tackle the fiscal policy difficulties precisely by establishing formal NRFs. The experience with NRFs goes quite far back; Kuwait’s General Resource Fund, for example, was established in 1960, and Kiribati’s Revenue Equalization Reserve Fund for phosphate revenues was established in 1956. Many currently active NRFs have been in existence for several decades – two examples considered relatively successful are the Alaska Permanent Fund (established in 1976) and the Norwegian State Petroleum Fund (established in 1990, although it did not receive any inflows until 1995). In the past years, there has been quite a scramble among natural resource producers to set up NRFs. Some of these are producers such as Sao Tome and Principe, which passed legislation establishing a permanent fund in December of 2004, while others are established producers, many of which have experienced political transitions in the recent past (e.g., Azerbaijan established its State Oil Fund in 1999, Kazakhstan set up its National Fund in 2001, and East Timor has set up a Petroleum Fund after the Norwegian model).
The stated purpose of these NRFs is to facilitate the accumulation of large, volatile, and temporary revenues when times are good; stabilise public spending, and finance public spending when natural resource revenues are no longer flowing in. The reality, however, is more worrisome. Studies show that it is difficult to detect a consistent improvement of fiscal policy in countries which NRFs relative to those without them. Fasano (2000) examined the NRFs of Norway, Chile, Alaska, Venezuela, Kuwait, and Oman, and found that the outcome is “mixed,” which reflects in part “the challenges in adhering to the operational rules” and the “overall fiscal discipline in the country.” Davis et al. (2003) carried out an econometric analysis of the effect of NRFs on the link between changes in public expenditures and variations in revenue. They found that while some countries with NRFs exhibit a lesser sensitivity of government expenditure to natural resource revenues than countries without NRFs, that advantage was already present before the countries set up their respective NRFs. There is, therefore, no evidence that adopting NRFs of these countries contributed to the soundness of their fiscal policies.
In practice, NRFs and similar institutions rarely make it any easier to accumulate large amounts of money in the orderly fashion the normative economic model calls for, reflecting the fact that the incentives surrounding the choices of politicians do not conform to this idealized model. For the fiscal policy to be correct, it is not sufficient to get the economics right. One must also get the political economy right, and as much as possible align the political economy incentives of decision-makers with what is good policy for the country.