Technical Change as a Stochastic Trend in a Fisheries Model

Right, I have a couple of things forthcoming. One is, as the post-title suggests, on technical change in fisheries, where I, in my first sole-authored paper in five years, suggest a state space approach to measure technical change in fisheries. The approach is applied to data from the Norwegian Lofoten cod fishery, a data set that previously has been analyzed with other, more typical methods (linear regressions).

The paper has a long history. It started in 2008, when I was a visiting grad student at the economics department of the University of California, San Diego (UCSD). There, Dale Squires, who I am proud to call my friend, presented an analysis of the Lofoten data. During my visit to UCSD, I had spent considerable time studying state space models and the Kalman filter, and during Dale’s talk I wondered whether a state space model would do a better job in estimating technical change. Dale’s analysis was published in 2010, at a time when I already had acquired the data and had started to develop a model and an algorithm. In 2011, during a train trip, I started to get promising results. Progress was doomed to be slow, however, because the entire project was a side project that I only worked on in short stints every now and then. At some point in 2012, I nevertheless had a manuscript ready for submission. I sent it to the same journal where Dale’s analysis was published. After an interesting and instructive review-process, the manuscript was rejected. In the years that followed, the manuscript was sent to a handful of journals (the manuscript took various forms over the years; condensed into the letter-format at one point), but the verdict was always the same: rejection. Over these years, Dale, who I kept in touch with, was always optimistic and encouraging, suggesting alternative journals. Early in 2015, the manuscript was finally sent to Marine Resource Economics, where it was accepted after no less than three rounds of revisions. In the last round, I had to pull out my initial version, written more than three years earlier, and add discussion that was revised out at some point along the road but which obviously had its place. The manuscript was formally accepted early this year (2016), eight years after I had the initial idea.

Late in 2014, more than six years into the process, I had another idea for how to carry out the analysis. I decided to pursue this new idea in another side project. This spin-off project had much faster progress, and less than six months later, a letter-form manuscript was already rejected. After some further work, expanding the manuscript to the more typical article form, the manuscript was submitted again, and I am now awaiting its review. This much faster progress on the second side project is partly taken, by me, as evidence that I have become better at what I do. The lower degree of complexity is, of course, also an important factor in the progress.

‘Technical Change as a Stochastic Trend in a Fisheries Model’ will appear in Marine Resource Economics during the fall. The abstract reads as follows:

Technical change is generally seen as a major source of growth, but usually cannot be observed directly and measurement can be difficult. With only aggregate data, measurement puts further demands on the empirical strategy. Structural time series models and the state space form are well suited for unobserved phenomena, such as technical change. In fisheries, technical advances often contribute to increased fishing pressure, and improved productivity measures are important for managers concerned with efficiency or conservation. I apply a structural time series model with a stochastic trend to measure technical change in a Cobb-Douglas production function, considering both single equation and multivariate models. Results from the Norwegian Lofoten cod fishery show that the approach has both methodological and empirical advantages when compared with results from the general index approach, which has been applied in the literature.

UPDATE: The article is now available here:
http://www.journals.uchicago.edu/doi/pdfplus/10.1086/687931
DOI: 10.1086/687931

Stochastically Induced Critical Depensation and Risk of Stock Collapse

My research team has published a paper in the latest issue of Marine Resource Economics (vol. 30, no. 3). The paper is called Stochastically Induced Critical Depensation and Risk of Stock Collapse and discusses how stochasticity induces depensation in fisheries models. We also develop a measure of stock collapse risk and apply it to a model with an optimal harvest rate. The abstract reads as follows:

This article investigates the risk of stock collapse due to stochastically induced critical depensation in managed fisheries. We use a continuous-time surplus production model and an economic model with downwardsloping demand and stock-dependent costs. First, we derive an optimal exploitation policy as a feedback control rule by applying the Hamilton-Jacobi-Bellman approach and analyze the effects of stochasticity on the optimal policy. Then, we characterize the long-term sustainable optimal state and estimate the risk of stock collapse due to stochastically induced critical depensation. We find that the optimal harvest policy in the stochastic setting is conservative at low stochasticity and approaches the myopic solution at high stochasticity. The risk of stock collapse is increasing with the stochasticity and decreasing with stock sizes.

Marine Resource Economics Impact Factor

Today, I discovered the impact factor of Marine Resource Economics is above 1. The MRE impact factor has only been measured since 2009. It started out in the territory around 0.5, which I found agreed well with my perception of the quality and standing of the journal. 1 is kind of a watershed, as I understand it, and the difference between 0.9 and 1.1 is more significant than the difference between 1.0 and 1.2.  Now that MRE is above 1, it is in the territory of journals like the American Journal of Agricultural Economics and Land Economics. It still has a lower impact, but less significantly so.

Fisheries Management Under Irreversible Investment: Does Stochasticity Matter?

In the latest issue of Marine Resource Economics (vol. 28, no. 1) I co-author an article together with colleagues at the Norwegian School of Economics. We explore what optimal management looks like when capital has to be managed as well as the fish stock. In fisheries, fleet investments are to some degree irreversible; fishing boats has little alternative use. While little has been written on the multidimensional decision problem of how large a fleet to own (the capital decision) and how much of it to use (the fishing decision), we also look at the effect of stochasticity. Things quickly become messy in these models as the number of potential scenarios is large. While it is manageable in our present case, I do not look forward to seeing further extensions. Or rather, I would love to see creative takes on analysis and presentations of high dimensional problems.

Our abstract:

We present a continuous, nonlinear, stochastic, and dynamic model for capital investment in the exploitation of a renewable resource. Both the resource stock and capital are treated as state variables. The resource owner controls fishing effort and the investment rate in an optimal way. Biological stock growth and the capital depreciation rate are stochastic in the model. We find that the stochastic resource should be managed conservatively. The capital utilization rate is found to be a nonincreasing function of stochasticity. Investment could be either higher or lower depending on the interaction between capital and the resource stock. In general, a stochastic capital depreciation rate has no strong influence on optimal management. In the long run, the optimal harvest for a stochastic resource becomes lower than the deterministic level.

Valuing Environment and Natural Resources

The two-volume tome Valuing Environment and Natural Resources, edited by Kenneth G. Willis and Guy Garrod, was published last year. It is interesting of at least two reasons. In the interest of backwardness, the least important reason is that it is titled almost identically to the book Valuing Environmental and Natural Resources (look again) by Timothy C. Haab (of env-econ.net) and Kenneth E. McConnell. As if the similarity in the title was not enough, they are both published on Edward Elgar.

The other, more important, reason for taking an interest in Valuing Environment and Natural Resources is buried deep in volume two. In chapter 32 of volume two, to be exact. Chapter 32, the first chapter in Part X (Marine), is namely a reprint of my very first published article. The article appeared initially in the journal Marine Resource Economics, volume 23, number 2, and was titled ‘The Premium of Marine Protected Areas: A Simple Valuation Model.’ The abstract goes as follows:

The article addresses the induced cost, the premium, from establishing a marine protected area in a deterministic model of a fishery. Outside the protected area, the fishery is managed optimally through total allowable catch quotas. The premium is found to be increasing and convex along the protection parameter. Biological measures are introduced to increase the understanding of the mechanisms in the bioeconomic system. Time-series solutions show that the net return per unit of fish increases after the protected area is established.

In fact, I discovered that the article was reprinted by mere accident (if to google yourself could be regarded a ‘mere’ accident). While I know the journal holds certain rights, I was surprised I was not even informed about the reprint. Oh well. Other authors who’s work are reprinted include big guns like Ian Bateman, Nick Hanley, John List, V. Kerry Smith, Jason Shogren, and a number of others and it is rather pleasant to be reprinted in the same book as them. The publisher describes the book as follows:

Over-exploitation of environment and natural resources is becoming increasingly widespread in the modern world. To combat this, environmental economists have attempted to value such resources in order to ensure that they are given due recognition in any ex ante appraisal, or ex post evaluation of projects or policies; and also to ensure that optimal levels of consumption are determined for the resource. This authoritative collection, along with an original introduction by the editors, brings together seminal papers published in the last three decades which demonstrate the application of a number of techniques employed to value a range of environmental and natural resources. It will be of immense value to students, scholars and practitioners with an interest in environmental affairs and natural resources. [Yes, italics are mine…]

 

More on World Fish Production

I just, presumably somewhat delayed, received the latest issue of Marine Resource Economics (Vol. 23, No. 4, 2008) . It’s a special issue on aquaculture (fish farming). There I found an interesting graph that relates to my earlier post on world fish production. The introductory piece (‘Aquaculture – Opportunities and Challenges,’ written by Asche, Guttormsen, and Tveteras) features a graph of world fish production from 1970 to 2006:

In contrast to the somewhat confusing graph from The Economist (see earlier post), it is clear from this graph that fisheries has leveled off since the late eigthies, while there is steady growth in farmed fish production. According to the article, farmed fish represented half of the fish produced for human consumption in 2008.

Since 1970, world fish productin has on average increased with roughly 3.5% per year and, from the graph, shows no sign of leveling off. To the contrary, farmed fish production has increased with approximately 12.5% per year since 1990! Indeed, aquaculture has been ‘the world’s fastes growing animal-based food sector during the last decades,’ according to the article authors. In comparison, the current global population growth rate is approximately 1.2% (see U.S. Census Bureau), and is expected to fall in the next forty years. Presumably, more and more people have started to, and will in the future, eat more fish.

Not all wild fish caught is used as food for humans, however; a substanial part of it goes into the production of farmed fish, for example. Further, more and more people eat more and more as developing countries moves out of poverty; world food consumption by humans will likely rise faster than the population grows. I’m not sure the authors have thought of these things, but they are anyhow optimistic on behalf of aquaculture’s growth prospects:

Given the status of global fisheries, with most large fish stocks being fully exploited or over-exploited, aquaculture production must increase in order to maintain or increase the global seafood supply per capita. Fortunately, the aquaculture sector seems well positioned to succeed in this respect.

Related posts:

• World Fish Production