One possible solution to the problem of overspending is to remove the government’s discretion over how natural resource wealth is spent. Most Natural Resource Funds (NRFs) are set up with conditions on how and when money can be withdrawn from them, with the intention of making the money in the NRF less easy to mismanage than revenues that are at the government’s free disposal in the normal budget. There are three types of rules institutional designers must consider:
1. Quantitative constraints: Rules governing how much can be spent.
An NRF can limit the government’s discretion over how much of natural resource wealth is spent by a rule or formula indicating the maximum expenditure allowed (or conceivably, the exact amount to be spent). In practical terms, the rules can express how much is available for spending either as a function of that year’s revenues, or as a function of total wealth, or some combination. Often rules are expressed as a function of the commodity’s price and its deviation from some benchmark level. Other times, rules limit spending to a proportion (typically the expected investment return) of the money already accumulated in the NRF, that is, a function of financial wealth rather than total natural resource wealth (i.e., excluding the resources not yet extracted). Clearly, limiting spending to a function of revenues is inferior from an economic point of view to limiting it to a function of wealth, in as much as a purpose of the fund is to divorce spending patterns from year-to-year revenue fluctuations. Even on this question, however, one must take into consideration how political pressures may act differently on the two types of constraints.
Regardless of how the ceiling on yearly aggregate spending is specified, the legal status of the rule can vary considerably. In fact, “rules” are situated on a continuum between discretion and rigid formulas. On one end of the spectrum there are “rules” that have no legal force and only reflect the government’s intended policies or policy “guidelines” or “commitments” (an example is the Norwegian State Petroleum Fund, and to a lesser degree the Petroleum Fund of East Timor). At the other end are formulas for expenditure ceilings enshrined in law (such as in Sao Tome and Principe), or with constitutional force (as in Alaska).
Depending on how mandatory a rule really is, restricting policymakers’ discretion comes at a cost of flexibility. Governments can respond less easily to crisis situations or unforeseen changes that change the optimal policy, but it can also give current governments confidence that the results of prudent expenditure in one period will not be squandered through lavish expenditures in future periods. Nevertheless, quantitative caps are probably not a sufficient solution to the basic commitment problem that affects policy choices in resource-rich countries. The core problem, as our analysis suggests, is that policymakers will have an incentive to overturn or ignore these caps, because the incentives to spend too much too early derive in part from concerns about how, rather than simply how much, the money will be spent.
2. Qualitative constraints: Pre-dedicated expenditures
In addition to codified rules for the amount of aggregate spending, an NRF could have rules for how the money is to be spent. Ecuador, for example, allocates excess oil revenues (above a budgeted oil price) to various funds with specific uses determined by law. For some of these, the decision of the aggregate amount to spend remains at the president’s discretion, but the allocation to various uses must comply with the law. Chad’s oil revenue management law provides for fixed percentage allocations of oil revenues to special “priority sectors” such as health and education. Alaska state law specifies that 50 percent of the investment return on the principal of the Alaska Permanent Fund be distributed to all the state’s residents on a per capita basis (the remainder has to be used to inflation-proof the principal before it can be spent on other uses). In other countries, fund rules state the general purposes for which the money should be spent, but in such a broad way that almost everything would seem to be allowed. The Oil Revenue Management Law of Sao Tome and Principe states the following: “The allocation of the Annual Funding Amount shall be decentralized with respect to sectors and territory, and aimed at the elimination of poverty and the improvement of the quality of life of the Saotomean people, the promotion of good governance, and social and economic development. In addition, such allocation shall be used, namely, to strengthen the efficiency and effectiveness of the State Administration, to ensure a harmonious and integrated development of the Country, a fair sharing of the national wealth, the coordination between economic policy and social, educational and cultural policies, rural development, preservation of the ecological balance, environmental protection, the protection of human rights, and equality among citizens before the law.” Again, the rules of existing NRFs vary widely in how legally binding they are for the country.
Proposals to impose such qualitative rules are commonly met with a concern that pre-dedicated expenditures may occasion parallel budgets, which lead both to a loss in allocative efficiency and to a decline in transparency. As a practical matter, steps can and should be taken to ensure that this does not occur, by requiring that pre-dedicated expenditures are entered into the general budget, or ideally, that qualitative constraints are referenced to the entire budget and not simply to the portions financed by an NRF (Humphreys and Sandhu 2007).
3. Rules governing inflows to the NRF
Many NRFs are set up with specific rules as to what should go into the NRF. This is not always necessary – money can be placed in the NRF by a discretionary act of the incumbent authorities when revenues are high. Discretion, however, allows governments to circumvent the rules the NRF imposes on spending by simply directing money straight to the budget without passing through the NRF. In the case of Chad, for example, signatory bonuses were not required to enter the fund with the result that initial expenditures of oil revenues in Chad were used for military rather than developmental purposes. Several countries accordingly channel all their natural resource revenues, including signatory bonuses, into their NRFs; others a fixed proportion. In some countries, NRFs receive some proportion of “excess” natural resource revenues, where the excess is defined relative to some budgeted price for the commodity in question (this is the case for Ecuador’s oil revenues and Chile’s copper revenues, for instance).
Benefits and Drawbacks of Rules for the Political Economy
How likely are rules to improve the political economy challenges discussed? They may do so if breaking the rule imposes costs on the policymaker, or if policymakers recognize that when they follow these rules, other policymakers may also be more likely to follow them. When policymakers engage in repeated interactions over a lengthy period of time, a reciprocating attitude may arise. When this happens, an incumbent will refrain from overspending today in the expectation that challengers who gain power in the future will do the same, since this will benefit the today’s incumbent when they at some point return to power again. If such expectations take hold, they may become self-fulfilling. NRF rules, even if they do not by themselves create incentives for restraint, may help generate such expectations and serve as a benchmark for identifying when one or another party has failed on their side of the agreement. The nature, magnitude, and probability of such costs will, however, vary both with the legal status of the rule (abandoning a policy commitment is different from breaking the law) and with the local political and institutional context. Even the authority of the law is a real problem in many countries, and a law that the government has an incentive not to follow may not be complied with.
This points to a deeper problem: Even if a rule stating that the government should not overspend changes the political calculation (because it is costly to break), it does not remove the original reason why the government wants to overspend – it merely counterbalances it. The original incentives may, therefore, prevail even in the presence of a legally enshrined rule. Just as we were earlier asking what incentives a policymaker had to deviate from the economically optimal policy, we may now ask what incentives a policymaker has to violate the rule (even one with legal force) that tells them to adopt that policy. Rules that implement spending paths that are not Pareto improvements on the status quo ante may be brittle, in the sense that they may be flouted at the first spike in commodity prices. Worse, they may also, by creating or reinforcing a precedent of rule-breaking, weaken the force of law overall. Designers of NRFs, therefore, must strike a compromise between creating incentives for the right policy and incentives to comply with the rules themselves.
Creating the incentives to comply with the rules is more than a legal issue or issue of institutional design. It also touches on aspects of investment policy, economic planning, and political communication. In some instances, for example, decisions regarding how monies in the fund are to be invested can affect the costs and benefits associated with ignoring, altering, or complying with rules. For example, fixed-term investments with a long term to maturity can in principle help enforce quantitative restrictions by reducing the liquidity of assets, although the impact of such investment policies on the incentives to alter or ignore the law will naturally depend on the country’s access to debt financing. In many cases, we may expect that the political costs associated with ignoring rules, such as rules allocating expenditures to key development areas or to particular regions, to be greater the more embedded the rules are in the longer-term development planning within the country and the greater is the information available to the beneficiaries of those rules regarding the targeted funding.