Syndicate content

How CGE models influence policy

Shanta Devarajan's picture

Sherman Robinson and I are writing a chapter in the forthcoming Handbook of Computable General Equilibrium (CGE) Modeling on the contribution of CGE models to policy formulation in developing countries.  The paper builds on our earlier piece.  As we did with that paper, we are seeking examples where CGE models have been used (or not been used) in policy discussions.  Please send them as comments on this blog.  Feel free also to comment on the proposed abstract for the paper:


“This paper reviews the experience of CGE models from the perspective of how they have—or have not—influenced public policy in developing countries.  The paper describes different classes of models, from stylized (“toy”) models, to large, multisectoral models to dynamic, perfect-foresight models, and shows what roles they played in shaping policy decisions.  The paper notes the evolution of these models, following different traditions in development economics, from central planning to neoclassical economics to structuralism to endogenous growth, and shows how the particular nature of the models affected their relationship to policy.  Finally, the paper looks at the evolution of policy formulation in developing countries, and shows how the need for country-specific information (rather than rules-of-thumb) and increased reliance on evidence-based policymaking affects the use and influence of CGE models in policy.  The paper distills the lessons from these three approaches to derive a set of characteristics that help determine whether, and how, CGE models will have an impact on policy formulation in developing countries.”

Comments

Submitted by Stephan Klasen on
Dear Shanta: Greetings from Germany and all the best for the nea year. In 2003 and 2004 I worked with people from the Kiel Institute on using CGEs for policy analysis in Bolivia, as part of the cross-country work on the drivers of pro-poor growth. The papers examined the rather limited ability to affect poverty via macro policy in a highly dollarized economy. While the results were discussed intensively with the government (and UDAPE), it is unclear that they had a direct impact on development policy making, as the government was changing to the MAS under Morales at the time. Here are three working papers based on this: http://www.uni-goettingen.de/de/67061.html, see papers 101, 120, and 143. Two publications that emanated from this: Klasen, S. Macroeconomic policy and Pro Poor Growth in a dualistic economy: The Case of Bolivia. In Cornia, A. (ed.) Pro-Poor Macroeconomics: Potentials and Limitations. London: Palgrave (2006), 305-325. Klasen, S. Melanie Grosse, Jann Lay, Julius Spatz, Rainer Thiele, und Manfred Wiebelt “Analyzing pro-poor growth in Bolivia: Addressing data gaps and modeling policy choices. In: Grimm, M., S. Klasen, und A. McKay (Eds.) Determinants of Pro-Poor Growth. London: Palgrave (2007), 191-218. I believe the Lay, Thiele, Wiebelt paper was also published in the Review of Income and Wealth. Best, Stephan.

Submitted by Dino on
Shanta - interesting review. As part of the work on economic growth and fiscal policy we used the MAMS CGE model with the Government of Uganda (you should consult officials also - they eventually ran the model for themselves) to consider what alternative growth paths would imply for the sustainbility of recent high rates of economic growth, and for fiscal strategy. A key conclusion arising from the general equilibrium perspective on growth was that agrarian Uganda would need even more rapid structural transformation and higher export growth to generate jobs outside of agriculture. Without the "double demand for food" produced in Uganda; from both foreigners (exports) and from workers in food processing export industries and services, the real price of agric products would fall more rapidly than the improvements in agric productivity needed to sustain growth with poverty reduction. In turn this growth path would be yet more urban and more infrastructure-intensive than Uganda's recent high economic growth, implying that growth constraints identified in our growth diagnostic work were likely to emerge if Uganda did not invest more heavily in urban transport and in energy infrastructure. Fiscal policy scenarios for financing the required increase in infrastructure threw up some interesting conclusions because of the MAMS link between budget composition and MDG production functions. These were intuitive but not immediately obvious in policy dialogue. First, since much of Uganda's health spending is private, an increase in taxes to finance a deficit caused by increased infrastructure spending (even though tax collection is low as a share of GDP) could reduce health outcomes relative to the status quo by reducing private health demand. Second, despite the assumed existence of Dutch-disease effects of aid, the benefits of relying on aid to finance infrastructure were preferable in growth and welfare terms to relying on taxes. (Bevan and Adam's CGE work for the CEM in fact concluded that Dutch disease effects of aid were offset by productivity-enhancing infrastructure, so long as gestation lags (between the receipt of funding and completion of projects) was minimized). Third, given that Uganda's budget had a relatively high share of "productive" spending, modest efficiency gains in education and health spending could yield higher welfare benefits than feasible cuts in "unproductive" spending. By extending the MAMS simulation period we considered the welfare returns to investments in family planning. (Uganda has the highest youth dependency ratio in the world and the world's fastest growing workforce, making it an interesting CGE case study of demographic dividend).

Submitted by Tehmina Khan on
Hi Shanta The CGE-MAMs model is being used for a number of purposes by the East Asia Prem team for Mongolia. For example, we have used it to model potential Dutch Disease impacts on Mongolia (which has large mineral resources and where the mining sector is being scaled up substantially over the next few year). The same model also underpins a policy note that focuses on raising female participation in the mining sector. In addition, as "short term analysis" we have used the model to illustrate to the government what we think will be the impact of a highly expansionary budget for 2011. If you want i can email you the related documents. Tehmina

Submitted by Francisco Carneiro on
Shanta, Great to hear that you are still working on CGE models and doing this review, which should provide interesting findings and insights for researchers and practitioners. I have contributed a chapter to a book edited by Rob Vos, Enrique Ganuza, Sam Morley, and Sherman Robinson (entitled Who Gains from Free Trade? Export-led growth, inequality and poverty in Latin America, Routledge Studies in Development Economics, New York, 2006) that used a CGE model to investigate the impact of trade openness on employment, poverty and inequality in Brazil (my chapter was entitled “Brazil – the impact of trade openness on employment, poverty and inequality” and was co-authored by Jorge Arbache). The book is a collection of articles addressing a similar question for a number of Latin American and Caribbean countries and presents the result of research funded by a joint UNDP-ECLAC research program that involved prominent think-tanks in several countries (e.g., IPEA in Brazil). So I think it is a good example for your review of how CGE models have been used in policy discussion. I have also published a paper with Jorge Arbache that used the CGE modeling approach in World Development that is frequently cited in academic papers dealing with trade openness in Brazil. The paper shows that trade liberalization in Brazil has contributed to improve economic welfare by means of greater output, lower domestic prices, and higher labor demand. However, one important finding of the paper was that the benefits of this economic improvement in Brazil have often been appropriated by the most skilled workers in the most trade-oriented sectors, contradicting the predictions of the Heckscher-Ohlin and Stolper-Samuelson theorems. The complete reference for this paper is: Carneiro, Francisco G. and Jorge S. Arbache (2003), The Impacts of Trade on the Brazilian Labor Market: A CGE Model Approach, World Development 31, 1581-1595. Good luck with this interesting work. Best, Chico.

Submitted by Olu Ajakaiye on
Dear Shanta, The piece by you and Sherman will certaily make a great contribution to the challenges of applying CGE model results in policy making. I partiuclarly like the focus on country cases as opposed to some large multicountry modles that tend to produce rules of the thumb type advise. Recently, Adeola Adenikinju, Bernard Decaluwe and myself edited a book containing a number of applications fo CGE models to policy issues in Nigeria. Most of the work was actually done by Adeola. i will ask him to forward a copy to you. The book contains some of my own works in this area and I hope you would find it useful in the piece. While looking forward to the volume, please accept my very best wishes. Olu

Submitted by Joaquim Oliveira Martins on
Dear Shanta, Such a Survey would be very much welcome indeed. In the OECD, CGE models have influenced the policy debate about global North-South issues, such as climate change. In the early 1990s, the GREEN model provided evidence on the differences in marginal abatement costs between OECD and developping countries, illustrating the potential cost efficiencies related to carbon trading. The model was also instrumental in predicting the increasing penetration of coal in primary energy demand of major emerging markets. Model's results also highlighted for example sensitive policy issues, such as the famous "carbon leakages" problem. Perhaps a even more influential role in shapping policies would have been obtained if the investment in this type of models could have been sustained over time. Structural modelling, like statistics, is a key input for the much wished "evidence-based policy making". Below are some references to the work with the GREEN Model. Best, Joaquim -Burniaux, Nicoletti and Oliveira Martins (1992), "GREEN: A Global Model for Quantifying the Costs of Policies to Curb CO2 Emissions", OECD Economic Studies, no. 19. -Oliveira Martins, Burniaux and Martin (1992), "Trade and the Effectiveness of Unilateral CO2-Abatement Policies", OECD Economic Studies, no. 19. -OECD (1995) Global Warming: Economic Dimensions and Policy Responses, OECD, Paris.

Dear Shanta et al., ( not sure if it is not too late make comments) I think this is going to be an interesting chapter. A lot of important things have happened in developing countries over the years and the contribution of CGE models to policy formulation since the earlier work of Dervis et al. (1982) of course has play a vital role in guiding policy makers in the scientific and effective way. As I have been working on my PhD project since 2008, I used a macro-micro economic framework by connecting the dynamic general equilibrium model to a Microsimulation model to examine the forestry policy change in Laos and this leads to two different objectives for my study. the dynamic general equilibrium model was used to examine impacts of forest plantation development on the macro-economy of Laos. Analysis is focussed on the Government of Laos’ forest policy involving the doubling of the 2003 forest cover by the year 2020. Some key conclusion from this analysis shows that this forestry policy is likely to have a positive impact on the Lao economy by increasing the production of forestry and forestry related industries and by stimulating exports and household income. Despite prompting a relatively higher growth in the production of various other industries, some undesirable impacts are also likely. There is only a fairly small increase in the growth of government revenue. Interestingly, when looking at the impact of this policy change at the household level. This policy change tends to have positive but minor effects on the households by increasing real household income and lead to reduction in poverty. Looking forward to reading this interesting chapter. Best regards, Somvang

Submitted by Drake Kyalimpa on
Dear Shanta, Thank you for the wonderful piece and insights on CGe modeling. The piece by you and Robinson will be worth reading. You have alreday contruibuted immensely in this subject area. Having a an econometrics background, I personally take great pride in the use of SAM-CGe Modeling framework in informing policy makers of the impact of their actions on the economy, more particularly on the microeconomy. We would not know or use traditional econometric tools to nalyse economy wide issues without CGE modeling. You continued enlightment in this area is much appreciated. I have analysed the Ugandan economy using the SAM-CGE modeling. I was particularly interested in the determination of key sectors and how different policies affect these sectors. My results indeed reflect the true nature of the Ugandan economy, especially the finding that the agriculture sector seems to be the most important sector for poverty alleviation and growth. I think it is fair to say that developing countries will be fairly served if they concetrated the limited resources to sectors that have significant linkages with the rest of the economy. Thiose sectors that spur growth while associated with high reditributive effects (i.e. high employment, household and factor incomes) rather than waste resources in non-productive sectors (sectors that do not create significatnt redistributive effects). I would like more insights on dynamic CGE modeling and how the finacial sector can be modeled within the CGE framework. There are not so many studies that have analysed the role of the finacial sector. I am particularly interested in a simple model that you might be knowing which could be the starting guide for financial sector modeling in the SAM-CGE framework. Looking forward to reading your article. Kind regards, Drake Kyalimpa

Submitted by Drake Kyalimpa on
Dear Shanta, Thank you once again for allowing us to comment on the role of CGE modeling. Please find the edited version of my article: Your article together with Robinson will be worth reading. You have already contributed immensely in this subject area. Coming from an econometrics background, I personally take great pride in the use of SAM-CGE Modeling framework in informing policy makers of the impact of their actions on the economy, more particularly on the microeconomy. We would not know or use traditional econometric tools to analyse the impact of policies on socioeconomic system without CGE modeling. Your continued enlightment in this area is much appreciated. I have analysed the Ugandan economy using the SAM-CGE modeling. I was particularly interested in the determination of key sectors and how different policies affect these sectors/agents. My results indeed reflect the true behaviour of the Ugandan economy, especially the finding that the agriculture sector seems to be the most important sector for poverty alleviation and growth 9sector with strong forward and backward linkages). I think it is fair to say that developing countries will be fairly served if they concetrated the limited resources to sectors that have significant linkages with the rest of the economy. Those sectors that can spur growth while generating significant reditributive effects (i.e. high employment, household and factor incomes) rather than waste resources in non-productive sectors. I would like more insights on dynamic CGE modeling and how the finacial sector can be modeled within the CGE framework. There are not so many studies that have analysed the role of the finacial sector. I am particularly interested in a simple model that you might address the modeling of the financial sector in the SAM-CGE framework. Looking forward to reading your article. Kind regards, Drake Kyalimpa

Add new comment