This week, I’m featuring a guest blog by my colleague, Dr. Jeffrey Vincent, the Clarence F. Korstian Professor of Forest Economics and Management at the Nicholas School of the Environment at Duke University.
—–
Most people would agree that it’s a good thing to improve governance: to make governments more accountable, bureaucracies more efficient, corruption less common, and property rights and the rule of law stronger. Most people probably wouldn’t think about the effect of improved governance on the environment, but people in the environmental policy community do, and they think it’s good, too. The forest sector has been the focus of most of this attention. Over the last decade, a raft of international organizations have launched initiatives to improve law enforcement and combat corruption related to forest resources. If you haven’t heard of these initiatives, it might be because many are under the banner of one of the worst acronyms ever written: FLEGT.
There’s empirical support for the idea that improved governance can reduce deforestation. In 2000, UC Santa Barbara economists Henning Bohn and Bob Deacon published a paper in the American Economic Review that investigated the relationship between deforestation and ownership risk in 62 developing countries. They found significant evidence that countries where ownership risk was higher lost forests more rapidly between 1980 and 1985 than countries where this risk was lower.
One of my hobbyhorses is the belief that deforestation is caused by logging. This seems logical—if you cut down trees, then there’s no forest left, right?—but it forgets the fact that a forest is a renewable resource that has the ability to recover from disturbance. Although logging can sometimes lead to the permanent loss of forest cover, in most cases deforestation results not from demand by loggers for trees, but rather from demand by farmers for the land the trees grow on. When I heard Bob Deacon present a draft version of the AER paper at a seminar at Harvard in the late 1990s, I immediately began wondering whether the results applied to logging. Does improved governance reduce timber harvests, and not just deforestation?
An even earlier (1985) paper by another University of California economist, Hossein Farzin at UC Davis, in the Journal of Political Economy contained results that suggested the answer might not be yes. One of the two most important ideas in resource economics is Hotelling’s Rule, which implies that a higher discount rate causes resource users to shift extraction toward the present—to accelerate resource depletion. (You’ll need to guess the other idea. Hint: think “Tragedy of the Commons.”) Farzin pointed out that while this was true for a user who had already made the investment required for extraction, the opposite could be true for a potential user still deciding whether to invest. In the latter case, a higher discount rate would reduce the present value of profits from resource extraction, and this could reduce investment. With less investment, there would be less extraction. So, a higher discount rate has opposing effects on resource extraction, and the net effect can be either positive (extraction increases, if the Hotelling effect is stronger) or negative (extraction decreases, if the investment effect is stronger).
What does this have to do with governance? Improved governance reduces risk, and reduced risk reduces the discount rate. So, we can expect improved governance to mirror the effect of a reduced discount rate and to have opposing effects on timber harvests. In 2010, a decade after the Bohn and Deacon paper (research takes time!), my former PhD student Susana Ferreira (now at the University of Georgia) and I tested this hypothesis in a paper, “Governance and Timber Harvests,” published in Environmental and Resource Economics. We compiled annual data on timber harvests, governance, and other variables during 1984-2006 for 67 developing countries. The panel structure of the data enabled us to control for unobserved differences across countries that could potentially confound the effects of governance. If you take Prof. Bennear’s course, ENV 350, “Program Evaluation,” you’ll learn how to do this.
Consistent with Farzin’s paper, we found that improved governance sometimes reduces timber harvests but other times raises them. Interestingly, it tends to raise them in countries with the worst governance. In those countries, where corruption is the most pervasive and law enforcement the weakest, investment in the equipment, roads, and mills or port facilities required for logging is evidently so depressed that it outweighs the Hotelling effect, with the net effect being to reduce timber harvests. Improving governance reverses this and causes harvests to rise.
This is a disconcerting finding for organizations that are promoting improved governance in the forest sector. It suggests that their efforts could wind up raising timber harvests in the very countries they are most concerned about. Now, higher timber harvests provide some important economic benefits, such as increased employment, government revenue, and foreign exchange earnings. These organizations believe, however, that timber harvests are already excessively high and are causing undue environmental harm. Our results suggest that cleaning up governments could make these problems worse.
This is not to say that improving governance is a bad thing. Improved governance provides other benefits, and our results imply that it can indeed be expected to reduce timber harvests in countries with relatively stronger governance. The important lesson is that policy interventions can have unintended consequences. If this is known in advance, then it might be possible to modify the interventions to manage those consequences. In the case of timber, such modifications could include more careful monitoring of logging investments and logging activity. Just improving governance is not enough.