On a university campus where productive research is the norm and students’ academic and professional pursuits are diverse and highly competitive, some degree of interaction between the research world and student life seems inevitable. While this manifests itself in a number of ways, one such instance was particularly noticeable to students who passed from the first to the second year of the master’s degree over the summer.
An opportunity for an experiment presented itself as students applied to their M2 track of choice and the directors of these degrees sought to piece together a program consisting of the strongest candidates possible. Historically, students who successfully passed the M1 year have been guaranteed a place in one of the seven M2 tracks; in most years, students ranked their top two options, and the M2 directors, working from these rankings and the capacity and prerequisite constraints for each program, negotiated and bargained until every student had a place in a program. As became evident to all parties involved, this allocation procedure induced a large amount of strategic and provincial behavior on the part of students and M2 directors.[1] At the urging of the administration and with the aid of some external funding, two TSE researchers, Yinghua He and Thierry Magnac, assisted by Christophe Lévêque and Adrian Torchiana, set out to design an experiment and examine the results of a number of allocation mechanisms, in turn, implementing the one which elicited a stable matching and the highest overall welfare amongst students and M2 directors. The literature covering the fields of mechanism design and school choice is rich and varied.[2] The high profile nature of matching programs like the National Resident Matching Program (NRMP) and the New York City High School Match (NYCHSM) program, and the awarding of the Nobel Prize to Roth and Shapley in 2012 has firmly placed mechanism and market design into the public conscious as something that economists regularly get right. Ultimately, if the NRMP can fill 96.4% of 29,000 residency openings from a pool of over 34,000 applicants, and the NYCHSM can match over 90,000 students to more than 700 high school programs with 83-85% of students ending up in one of their top 5 choices, mechanism design is alive and well.[3] For TSE, with only 150 students, 7 programs and a handful of prerequisite constraints for admission to certain programs, theory could be expected to have no problems transitioning into practice. The TSE experiment collected student preference rankings under four mechanism schemes: the student-proposing Deferred-Acceptance (DA) Mechanism (where students rank-ordered the seven M2 programs), the Boston Mechanism (again, with students ordering all programs), the DA mechanism with students applying to only their top four programs (thus, potentially reducing review costs), and the DA mechanism with students applying to their top three programs, with the opportunity to apply for further programs at the cost of writing a letter of motivation for each additional program (again, potentially reducing review costs).[4] Students were told that one mechanism would be chosen randomly, and that the results from this mechanism would be used in the allocation. Initially, this multistep procedure was met with criticism by a large number of students and some M2 directors. Some found the procedure needlessly complex and confusing, while others thought experimentation was being performed at the expense of student welfare. However, it seemed that as people became more comfortable with the theory driving the experiment, they saw the present and future value in improving the mechanism. The DA mechanism is known as a strategy-proof mechanism, meaning that one side of the market (here, students) can optimize their welfare by honestly revealing their rankings.[5] On the other hand, the Boston mechanism has been shown to be manipulable, meaning it may reward strategic misrepresentation of preferences.[6] Similarly, due to choice restrictions placed on students by the DA mechanisms with constraints and motivation letter costs, the mechanisms may provide incentives for students to rank-order program preferences strategically. Thus, it would be expected that sophisticated students provide different rankings for different mechanism procedures if there are strategic advantages to doing so. Ideally, if the correct mechanism is chosen, a stable matching - one with no potential Pareto improvements (on the student side, but also on the program side) – is found, and students end up in a program that corresponds to their motivation and academic interests, thus improving academic performance and therefore boosting student welfare, net of costs induced by the mechanisms. On the other side of the market, directors were encouraged to rank all 150 students. Due to review costs, however, it was deemed more practical to supply each director with a “focus group” of around 50 students that had a high likelihood of being assigned to their program. These sets of students were generated algorithmically from a preliminary test round of the mechanism based on grades, program constraints, and preliminary student preference reports. Thus, each director did not need to review and rank each of the 150 students participating in the allocation, but instead, benefitted from a reduced burden of review costs. In aggregate, review costs can be massively expensive in the sense of time lost- this development signifies a large welfare gain on the part of M2 directors. Once these two sets of preferences were constructed and submitted, an algorithm designed and executed by the research team was run, and results for each of the four mechanisms were obtained. The overall results of the experiment were in some ways expected, and in others, rather surprising. Based on the stability of the match, TSE Director Jean-Philippe Lesne’s preference for a non-strategic mechanism, and the criteria used to analyse student’s welfare ex-post[7], the standard DA mechanism was deemed to be the best mechanism. Overall, 91% of students were placed in their 1st choice program, 7% placed in their 2nd choice program, and the remaining 2% placed in their 3rd choice program. In comparing these results with the outcomes from the other 3 mechanisms, marginally different percentages were obtained in terms of allocation to each person’s 1st, 2nd and 3rd choice programs, but in no mechanism did a student find himself allocated to their 4th choice program. Perhaps more surprisingly, the results showed a greater than 90% correlation between students’ preference rankings under the different mechanisms. Most students did not utilize strategic behavior when ranking under the Boston and constrained DA mechanisms. With the knowledge that all participants were fairly sophisticated[8], and well aware of the rules of the mechanism and the potential advantages of ranking strategically under manipulable mechanisms, the reason for this high level of correlation is not superficially evident. Perhaps the time and effort cost of calculating the optimal strategy for each mechanism was deemed too high for students. Perhaps in some cases, the strict ordinal student preferences captured by the ranking did not adequately measure the intensity of students’ preferences.[9] Perhaps the fierce competition for a couple of programs and nearly open-ended admission to others meant that students were willing to take a risk and state true preferences under the Boston mechanism.[10] In any case, strategic play was quite rare, and ultimately, with the utilization of the DA mechanism, proved irrelevant. Looking forward, with the positive feedback about allocation results that the administration has received this year from students and faculty alike, students should expect a similar DA mechanism to be used for allocation next year. Citing the proposed changes to M2 doctoral track admission for next year, Mr. Lesne tentatively believes that a two-step procedure will be used either on the student side or the program side, whose aim would be to allocate first students applying to the doctoral M2 (ecomath), then use a DA mechanism to allocate the remaining students to the six professional M2 programs. M1 students should anticipate a meeting to discuss the allocation procedures towards the beginning of the spring term. Footnotes: [1] Consider, for example, a weak student who is most interested in gaining acceptance to the EMO master, one of the more competitive programs. Being required to submit only 2 preferences, by listing ECOMATH- the most competitive M2 track- as his top choice (for which he will surely be rejected on the basis of grades), and listing EMO as his 2nd choice, he has thus put the M2 directors in an awkward position, where if the student is not admitted to the EMO, the remaining directors have little further information as to the student’s academic preferences. Thus the EMO director can admit him (despite his weaker grades) or the student can be randomly (or grudgingly, following bargaining/negotiating amongst directors) allocated to one of the 5 remaining programs, both of which are suboptimal, inefficient results. [2] For a short overview of mechanism design, see the excellent Nobel Corner summaries written by Yinghua He, Michel Le Breton and Jerome Renault in the third issue of the TSEconomist. For a more detailed review, see Pathak, 2011: http://economics.mit.edu/files/6390 [3] For this year’s report on the NRMP, see: http://www.nrmp.org/2013-results-and-data-book-press-release/. For more information on the NYCHSM program, see: http://schools.nyc.gov/NR/rdonlyres/58C1D35F-33DB-44AA-9C2A-896E000BD404/0/2013CitywideFairOverviewSession_Posted.pdf [4] Relevant descriptions of the mechanisms’ timings and processes can be found in the chapter titled “School Choice”, written by Abdulkadiroglu in the recently published (2013) Handbook of Market Design. [5] Roth, 1982, MoOR. [6] Abdulkadiroglu and Sonmez, 2003, AER. [7] In this case, the two main criteria used seem to have been a minimisation of students allocated to their 3rd and 4th best preferences and a maximisation of students allocated to their 1st best preference (assuming preferences under the truth-inducing DA mechanism). [8] Given their status as graduate students in economics and probable exposure to some degree of mechanism design. [9] For example, consider the student who would only stay for the M2 at TSE if they were admitted to a particular program. Otherwise, they are essentially indifferent between the other six programs. In this case, strict ordinal measures do not account for the intensity (or here, indifference) of this student’s preferences. Perfect correlation of his preferences across the 4 mechanisms makes perfect sense. [10] Here, imagine a student whose top 3 preferences are respectively, EMO (highly competitive), PPD (less competitive) and ECOMATH (highly competitive). Under the Boston mechanism, a student might fear listing these true preferences if he believes he will be rejected by the EMO in the 1st round, that there is a chance he will miss out on a place in the PPD in the 2nd round, and that he will certainly have missed out on an ECOMATH place in the 3rd round, thus meaning he will be allocated to his 4th round choice. At TSE, however, this seems a scenario of low probability due to course sizes and prerequisite thresholds for less competitive programs, and thus, the student will probably simply play their truthful strategy.
0 Commentaires
Questions: 1) You served as a member of the Bank of England Monetary Policy Committee during the onset of the sub-prime crisis and the consequent financial crisis that is known as “the great recession”. Could you describe us your feelings from that, we suppose, intense experience? Would you like to share with us an incident that describes how crucial that period was and how important the decisions were that you had to make under such time pressure? It was a learning experience. Academics are rewarded for having something important or original to say. Policy-makers are rewarded for getting it right, or at least, appearing to get it right. And academics can take their time. The immediacy of the crisis gave no time to learn, no time to properly understand what was going on. It was necessary to act at all costs using first principles. As far as monetary policy went, things were relatively straightforward once the crisis was in full flow. At least, the direction of policy and the need to anticipate what would happen at the lower-bound for interest rates became the focus of concern. But both sides of the crisis showed the limitations of monetary policy. Tightening policy prior to the crisis would likely have had little impact on what happened subsequently. The build-up of debt and the risks taken on by banks had much stronger driving forces than low interest rates. Equally, although it made sense to cut interest rates rapidly and to start the program now known as “quantitative easing”, this was never going to be a magic bullet. I will never forget participating in the coordinated rate cut in October 2008. The logistics were fascinating from the inside but also to see how events galvanized coordinated policy action was a real measure of the determined nature of the policy response. 2) Having experienced both sides of the life of a distinguished economist, the life in academia and the life in policy making in prestigious institutions, could you tell us which of the two you prefer? What are the advantages and disadvantages of each position? The easiest question you will ask. I am delighted to be back fulltime as an academic. And being away only reinforced the benefits of academic life. Best of all is having the time to think and write. And getting the constructive and collegial feedback on what you are doing from a range of interesting and intelligent people is a joy. But, of course, your sphere of influence is limited. I do like to have people read and even criticize my work. But the immediate impact is less clear. The role I played was as a policy maker, not a policy advisor, and that gave me a chance to have a direct impact on policy. Being in a position like that that can be seductive. But it is humbling too. Academia allows you to try to plough a deep furrow and to focus on creating a long-term understanding of issues. 3) What do you think about the policies that have so far been applied for the recovery of European countries from the debt crisis? Could you comment on the austerity-growth debate? I have throughout taken the view that we need to worry about getting the public finances back on track while pursuing growth. There are so many things that would make a difference to growth which can be pursued and would make a long-term difference. That is why I established the the LSE growth commission (with my colleague John VanReenen) and we have had some traction with some of the policy prescriptions that we put forward. I was not (and am not) in favor of saying that austerity should be postponed to some ill-defined point down the line – there are many public programs where there is a need to take a long-hard look in cost-benefit terms and the political will to tackle this will be greatest when the mood music favors austerity. And I strongly favor having a carefully spelled out and credible fiscal plan – with independent verification alongside a series of measures to support growth. But the balance has to be sensitive to developments in the economy. The problem with the weaker economies in Europe is that they have effectively lost their economic sovereignty. That is not only humiliating but it takes many of the important decisions out of the hands of your own elected politicians which undermines the democratic process. So it is important to stay clearly in control of your own destiny. 4) Do you believe that the political regime of China has played a vital role in its high economic growth? To what extent is there a place for autocracy in the industrialization and economic growth of poor countries? China faces many challenges and their leadership is aware of this. The growth and poverty reduction that have been achieved are truly remarkable in the sweep of history. China is an autocracy but there are a number of accountability mechanisms in place through the decentralized structures and the way that the Communist party operates. These explain their success in my view. But at some point, it will have to embrace the fact that citizen demands for a more pluralistic political system are likely to become overwhelming. Torsten Persson and I have been arguing that China should anticipate this by strengthening executive constraints, the rule of law and encouraging greater transparency. These are directions of travel but the optimal timing will be very tricky. And apart from general guidance, I don’t think that academic research will be much help in the finer-tuned aspects of this. 5) Your research has a focus on the conditions for a peaceful equilibrium within countries. How do you see the future progression of political and economic institutions in China given the rapid increase in living standards and education? Do you believe reform will be inevitable, particularly in the context of decreasing growth figures? There is no unconditional answer – it will depend on how the kinds of reforms which I outlined above progress. And there will be some luck involved – since there is vulnerability to a variety of potential shocks. The slowdown in growth is inevitable but how far that will be the result of convergence in living standards or citizens deciding consciously to trade off higher incomes against other aspects of the quality of life is hard to know a priori. 6) Your research work suggests that in U.S., political competition is good for economic performance. Does this result apply to other countries with different political systems? Can political competition give rise to corruption and rent seeking behavior of political parties that try to beat their competitors at any cost (bribes, lobbying, and use of their political power to help their supporters)? I am a big believer in political pluralism for lots of reasons. Societies that promote free expression are more open/creative and successful in many ways, and these are ways that are not easily captured solely by income levels or any single indicator. But an effective plural and competitive system has a set of underlying constraints to establish a core set of common interests/values. This is what Torsten Persson and I argue is promoted by cohesive institutions in the framework that we developed in out book Pillars of Prosperity. And this is what prevents competitive systems descending into disruptive partisanship where one group or another behaves destructively towards the other when in office. So you need the safe-guards against corruption/rent-seeking separately as the basis for having effective competition. A monopoly on political power is at best a second-best solution when constraints are weak – which might explain why China can work (for now). 7) It is clear that in policy debates the mass media play a crucial role in the determination of public opinion. Do you believe that during an economic crisis, when some immediate but harsh policy measures need to be taken (like Southern European austerity packages), media capture by the government (in favor of these measures) can lead to better political and economic outcomes? In what way should the state guarantee the freedom and the objectivity of the mass media? You can see an emerging theme. Silencing the media can never be first best in my view. Difficult economic circumstances call for efforts to create broad-based coalitions to make tough and reasoned decisions. There is no need to silence the media when there is enough consensus in the system. But I would also be pretty wary of silencing the media or allowing capture in a polarized society. After all, it is not going to be a social planner making the decision! I pretty much line up with Thomas Jefferson who said “If I had to choose between government without newspapers, and newspapers without government, I wouldn't hesitate to choose the latter." So I would place the burden of proof pretty high and try to get on with building the kind of political consensus on important dimensions of policy if it can be done. And that should help to provide the basis for tough decisions. 8) In your paper with Torsten Persson, “Repression or Civil War?” you conclude with the following sentence: “The ultimate goal is to map political and economic circumstances into our wider understanding of the forces that shape economic and political development”. Particularly in the context of political violence, to what extent can this wider understanding move beyond a theoretical structure and feed through into the active policy sphere? There are narrower and broader concerns here. The narrow ones involve looking at the specific factors which shape the use of violence in a particular context. Obviously there are tough debates about external intervention which we are seeing now in Syria. And there are examples which have been as broadly successful in Kosovo and Sierra Leone. But the wider issues involve thinking about how the institutional reforms can be put in place to create the kind of cohesiveness that is needed to build sustained peace. A concrete policy lesson, in line with much else that I have argued above, is that constraints on executive power and the rule of law are a priority ahead of running elections. Of course, we are not so great at knowing how to do this whereas we have a framework for conducting elections and monitoring their conduct externally. But just because we know how to do it, does not make it the first priority. So there is a practical research program on creating and sustaining checks and balances which would be a better focus. 9) When comparing development trajectories, economists tend to focus on economic factors like material and human resources, technology, economic institutions, etc. Is it a message from political economy that political institutions are at least as important as these other factors? If so, could you illustrate your response by an example? There are many. My reading of UK history would give a central place to series of judicious political reforms from Magna Carta onwards that created the conditions to sustain economic prosperity. In line with my argument above, these constraints on executive power existed before open elections came onto the scene in the nineteenth century. Other examples include the trajectories of Eastern Europe before and after the fall of the Soviet Empire. And surely even an arch skeptic would have to buy the argument when it comes to comparing the fates of South and North Korea. 10) Mechanism design has demonstrated its success in the realm of auctions and matching markets. Do you think that it can be applied with as much success in the design of political mechanisms (electoral rules, legislative bodies and so on)? It is a powerful way of thinking. But it too easily suggests that fine tuning is what matters. Getting basic things right should be what matters and can have a first order effect. So I don’t think that whether we have differences in the details over the rules for legislative organization matters that much even though there is a fascinating set of debates to be had about such issues. What matters is that we have legislators in the first place and that they have the power to act to promote good ideas and stop bad ideas when the need arises. That is the real achievement of Parliamentary democracy. Of course that is perfectly consistent with a mechanism-design perspective, it is just a bit less intellectually refined. Maybe that was something I have become more confident about as a result on having spent time in the messier policy world 11) Could you describe to us your role as a member of the Scientific Council of TSE? How do you see TSE developing over the coming years? My role is simply to try to provide advice from an external input – maybe some of it based on things that we have done right (and wrong!) at the LSE. I have much optimism. The vision of Jean-Jacques Laffont remains an inspiration and the power of his legacy is plain to see. You are fortunate indeed in having such dedicated people who lead by example and inspire. TSE is already one of the leading centres in Europe and from one SE (LSE) to another (TSE), competition is a good thing – it will keep all of us on our toes! 12) In your view, what are the next big questions to answer in the realm of Political Economy and Development? There are so many. But one thing that we need to understand better is the framework of norms within which institutions and economies operate. The recent experience of Egypt seems in part to be due to a failure to create a sufficiently strong ethic of common interests within which democratic institutions operate. This is not only down to formal rules – it requires that people accept and live by values of tolerance and decency. Women were poorly treated less than 100 years ago, being denied the right to vote and economic opportunities. The subsequent transformation in many parts of the world serves as a reminder that changes based on shifts in norms can be transformative. So I am convinced that the interdependence of changing norms, institutions and behavior are the key to progress in economic development and in society more generally. If we could understand that properly, it would definitely count as progress! To live is to take risks: Danger is all around us and the most we can do is to slightly reduce our exposure. Certain actions such as buckling your seat belt when driving or drinking only bottled water do in fact reduce the risk of incurring mortal injuries or contracting painful diseases; however, even such simple measures are not costless. Is a morning coffee worth the risk of hot water burns or the risk of spills on the laptop? Every moment of every day, we are making decisions that trade risks against some benefit we incur by accepting them. In my thesis, I attempt to deepen our understanding about how people value certain risks, and shed light on how government policies might nudge individuals towards better choices.
If we could and wanted to protect ourselves against identified risks, we would need to incur costs. Unfortunately, we have limited resources, thus, we need to find a system that allows us to minimize the expected impact that those risks may have at the lowest cost. Comparing diverse risk-reduction strategies requires a common metric. One popular approach is to use a monetary metric which facilitates the comparison between costs and benefits. Costs are usually expressed in monetary terms and are relatively easy to measure; generally they are computed using process-based calculations. Benefits are more problematic. In many cases, population-level risk reduction results in probabilistically saving the lives of unidentified individuals. How much is such a thing valued by society? The conventional metric for valuing a reduction in mortality risk in a monetary form is the Value of a Statistical Life (VSL). While the VSL is often misinterpreted, it does not purport to represent the value of an identified individual’s life. On the contrary, it is a measure of how much society is willing to pay to reduce a diffuse but possibly mortal risk. It would be valuable, from a policy perspective, to know exactly how much an individual is willing to pay to reduce a mortal risk to him or herself or to others; unfortunately, we hardly ever do. It is therefore necessary to try to extract this information from the decisions that people make. There are two main ways that this is done in practice: either through revealed or stated preferences. Neither approach is perfect, but both are powerful tools that allow researchers to estimate true preferences. On one hand, the main advantage of the revealed preference approach is that it is based on what consumers are actually choosing. Unfortunately, revealed preference measures face the critique that the effect that is being captured could possibly be confounded with other effects. Identification of the willingness to pay (WTP) to reduce risk is compromised. On the other hand, stated preferences overcome the issue of identification by controlling the decision-making environment but stated preferences are stated; we do not know whether the respondents would really behave as they have stated . Nevertheless, under the right circumstances, these tools can be valuable aids for evaluating or designing policies. VSL models assume that people make decisions based on their individual preferences. What about the risks that we take for others? Empirically, individuals tend to be willing to pay more for risk reductions relating to children. Evidence of this is can be found in the Food Quality Protection Act in the United States of 1996 which requires an additional tenfold margin of safety for children to ensure that they face no risks from pesticide residue in food (Dockins et al. s 2002). Why is this the case? First, while individuals tend to prefer risks that are voluntary (Slovik, 1987), children are generally perceived as involuntary participants in risky activities. Second, there is ambiguity related to the lifetime health risks faced by children, particularly for new or modern threats. Theoretically, an increase in WTP for risk reduction could plausibly stem from ambiguity aversion (Alary et al. 2012). Finally, there is some evidence to suggest that age could affect WTP (Rowe et al. 1995). In general, we assume that parents have the right incentives to care for their children, and therefore accept parents' valuations of their children's health. Of course, this is under the assumption that parents always have the information they need to take the right action, but do they? In my first paper, I use a revealed preference approach to explore a particular risk-reducing action taken by mothers. In 2000, the French government, following a worldwide trend, began a health advisory policy urging French residents to improve their eating habits. The policy, which is still being implemented today, is commonly known as “Manger Bouger”. Embedded within this larger policy, a smaller, lesser known initiative (starting from 2005) targeted fetal neural tube diseases (NTDs). NTDs are potentially deadly conditions which occur when the neural tube is not fully covered by the spinal cord. While NTDs often lead to abortions, the condition is generally not terminal; the consequences for the child include paralysis or severe brain malformation. To reduce the risk of NTDs, mothers need to consume at least 400 micrograms of folic acid (also known as Vitamin B9) on a daily basis for two months before and two months after conception. This amount can be achieved either through naturally occurring folic acid or through supplementary pills. Unfortunately, the NTDs trend did not change after the policy. Over the past decade the yearly NTD prevalence was roughly 1 baby per 1000. Does this mean that the policy did not have any effect? Using a highly detailed household level purchase database, a quasi-experimental setting and state of the art demand estimation techniques on the ready-to-eat breakfast cereal market, my research suggests that targeted women did in fact consume more folic acid after the policy was implemented. This increase was achieved through pill supplements. Regrettably, timing is everything. Supplemental folic acid taken outside the narrow time window will have no effect on fetal NTD risk. Although targeted individuals did consume more folic acid it may seem that they did not consume it at the appropriate moment. This is not surprising since it is very hard to correctly predict the timing of conception. Is there anything else that can be done? Fortifying staple foods is a common practice in France, it is an inexpensive and effective process. Nearly all the baguettes consumed are fortified with some vitamins. However, vitamin B9 is not among them. This omission is due to the plausible secondary effects that B9 can have on individuals aged 50 years and over. There are epidemiological studies linking increased levels of folic acid to the proliferation of some types of cancer. Others studies find an association with decreased levels of cancer proliferation. The bottom line: there is an important level of uncertainty regarding the secondary effects of B9 in some segments of the population. To deal with this uncertainty, I construct a probabilistic model and evaluate the impact of a massive B9 fortification policy in France: an increase in folic acid intake of 400 micrograms or more by the entire population. After taking into account the effects on longevity, health and wealth for children and adults, I conclude that a fortification policy is advised. In the second chapter of my thesis, in joint work with James Hammitt, we conducted a French representative Internet based survey to assess the WTP to reduce risks of fatal disease. The survey was designed to identify how WTP varies with characteristics of the disease (cancer or other diseases, which organs are affected etc.), the latency between exposure and the manifestation of the symptoms (1, 10 or 20 years), and whether the person at risk is the adult respondent, a child or another adult in the respondent's household. We use a latent class estimation technique along with paradata - data on how the survey data was collected - to identify those respondents who have correctly answered the survey. What is considered as answering a survey correctly? What is usually done in the literature is to check if, at least, the respondents are paying attention to characteristics that are hard to grasp. One of those characteristics relates to scope sensitivity. It is conventional that a survey passes the scope sensitivity test if the amount willing to be paid is nearly proportional to the risk reduction. We find that the proportion of respondents that are paying attention to these hard characteristics varies between 20 and 40 per cent. Time spent filling the survey influences heavily the quality of the answers: too little or too much time spent has a negative effect. Moreover, the implied VSL with and without our estimation procedure varies substantially: results range from 6, 10 and 8 million euros per statistical life for respondents, child and other adults respectively, based on the standard technique to 2.2, 2.6 and 1.8 millions euros with the latent class estimation. Although these results are still preliminary, they suggest that the differences in VSL we initially observed are partly explained by respondents answering the survey incorrectly. As compared with the US where the implementation of cost-benefit analyses to select the best projects is the norm, their use in France is still quite restricted. In part, this has limited the scope of WTP studies in France particularly regarding WTP for child risks. There are clear intentions from the French government to begin using cost-benefit analyses more frequently. This development highlights the need to investigate whether established policies are working, as well as to develop reliable French VSLs estimates for ex-ante (or ex-post) project evaluations. The Keynes-Tinbergen Debate on the Relevance of Estimating Econometric Models for Policy Analysis11/9/2013
Tinbergen’s motivation and basic approach to econometrics The desire to combat the socio-economic consequences of the Great Depression of the 1930s was Tinbergen's motivation for using econometric modeling. His approach towards studying periodic economic upswings and downswings contrasted with previous approaches to business cycle research. After a 19th-century undertaking by Juglar (1862) ascribing the recurrent business crises in Europe and North America to credit crises, and Jevons's (1884) study pointing to agricultural production cycles connected with sunspot numbers, several research projects in the early 20th century were devoted to the construction of so-called business cycle barometers. The purpose was to measure economic fluctuations through a particular index (or set of indices) with the aim of giving warning signals for turning points that would lead to a depression. An example was the Harvard Index of Business Conditions, known as the Harvard Barometer, constructed by a team led by Persons (1919). Another well-known descriptive approach to the business cycle during this period was initiated by Mitchell (1913). Mitchell’s work was followed by that of Yule (1927) and Slutzky (1927), who suggested that the cumulative effect of random shocks could be the cause of cyclical patterns in economic variables. Frisch (1933) applied these ideas, introducing econometric models in which impulse propagation mechanisms led to business cycles. However useful it could be as a starting point, Tinbergen criticized descriptive analysis as being too vague for use in policy preparation, and started a quantitatively oriented research program to explore the possible economic causes of the periodic upswings and downswings in economic activity. In an earlier theoretical study, Aftalion (1927) had argued that lags in an economic model could generate cyclical variation in economic activity. Following this argument, Tinbergen specified a first simple case using a system of difference equations to express lagged responses of supply to price changes in a market for a single good. He noted that the systematic fluctuations that could arise in such a system had been observed in an empirical study of the pork market by the German economist Hanau (1928), a phenomenon that became known as the ‘cobweb model’. Tinbergen subsequently generalized the specification of dynamic equations with lagged adjustment processes to macroeconomic settings, arguing that fluctuations in components of national product, such as investment and consumption expenditures, would lead to business cycle fluctuations in general economic activity. In 1936 he published the first applied macroeconometric model for the Netherlands. It was a dynamic model consisting of 22 equations in 31 variables. Employing what we now see as basic statistical techniques like correlation and regression analysis, it was to be used for the analysis of the particularly pressing unemployment problem. The specification of consumption and employment in this model anticipated elements of Keynes's theory (1936). This modeling exercise resulted in a strong policy recommendation in favour of a devaluation of the Dutch guilder to tackle unemployment. But its importance for the economics profession was far more profound: for the first time, the economic-policy debate had been based on empirically tested, quantitative economic analysis and not on rather informally stated economic theory, the so-called verbal approach. Thus, according to Solow (2004, p. 159), Tinbergen's work during this period ‘was a major force in the transformation of economics from a discursive discipline into a model-building discipline’. The Keynes-Tinbergen Debate The formulation of certain relations in Tinbergen's 1936 model showed some resemblance to Keynes's theory. Nevertheless, in an article in the Economic Journal of 1939, Keynes was remarkably skeptical of Tinbergen's work. Keynes labeled Tinbergen's method of estimating the parameters of an econometric model and computing quantitative policy scenarios as ‘statistical alchemy’, arguing that this approach ‘… is a means of giving quantitative precision to what, in qualitative terms, we know already as the result of a complete theoretical analysis’ (Keynes, 1939, p. 560). Their widely diverging views on the relevance of quantitative economic analysis were also illustrated by Keynes's reaction to Tinbergen's estimate of the price elasticity of demand for exports. When, in 1919, Keynes had strongly criticized the excessive war indemnity payments enforced upon Germany after the First World War, his argument had depended critically on the value of this elasticity. Tinbergen empirically found this value to be minus 2, precisely the value that Keynes had assumed a priori in his study. When informed about this Keynes replied: ‘How nice that you found the correct figure’. Keynes' critical attitude towards macroeconometric modeling and analysis originated from his view that the underlying economic theory should be complete in the sense that it should include all relevant variables and set out in detail its causal and dynamic structure. Econometrics could be used only for measuring relations (‘curve fitting’ was the term used); it could not refute economic hypotheses or evaluate economic models. Tinbergen, on the other hand, argued that economic theories cannot be complete. Econometric research could be useful for scrutinizing elements of economic theories and for examining whether one theory describes reality better than another. Further, it could provide the numerical values of the coefficients in dynamic models that determine the cyclical and stability properties of the model, and, by applying a testing procedure of trial and error, it could yield suggestions for an improved specification of dynamic lags. The debate still goes on In this controversy Tinbergen's approach soon gained the upper hand as increasing numbers of economists, especially in the United States, noted its practical results in terms of model construction and verification, including forecasting and policy recommendation in particular for monetary policy. However, Keynes's comments on the role of expectations and uncertainty in macro-econometrics and on specification and simultaneous equation biases remained relevant. Haavelmo (1943) advocated the use of probability theory in bridging the gap between theory and data in business cycle analysis. Later these issues would become the subject of intensive debate and research. The pioneering work by Thomas Sargent and Christopher Sims, Nobel Laureates in Economics in 2011, to construct models that both fit the data and can be used for forecasting and policy is a clear example of the continuing debate. Discussing their contributions is beyond the scope of the present note. I only mention that forecast and policy implications based on their modeling and inference techniques are studied and used by almost all econometricians at the US Federal Reserve System and at European Central Banks. Personal note I met Jan Tinbergen late in his life when I was director of the Tinbergen Institute, www.tinbergen.nl , from 1992-1998 and after his death from 2008-2010. A brief story of my personal experience with Tinbergen shows his interest in that empirical econometric work which may have enormous potential policy implications. In 1994, just a few months before he died, he read my paper on the bimodal distribution of the World’s Income and the very low estimates of the catch-up probability of poor countries with the rich ones, see Paap and Van Dijk (EER, 1998). He saw it as a testimony to his efforts to make development programming of poor countries an important area of theoretical and practical research and he invited me (with my wife) to discuss in more detail my paper on a Saturday afternoon in the Spring of 1994. Regrettably he passed away on the Monday before our meeting, but it shows how actively involved he was in following empirical econometric research that has substantial policy implications until the very last days of his life. This note is an adjusted excerpt from the paper: “Tinbergen, Jan (1903–1994)." By Cornelisse, Peter A. and Herman K. van Dijk, in The New Palgrave Dictionary of Economics. Second Edition. Eds. Steven N. Durlauf and Lawrence E. Blume. Palgrave Macmillan, 2008. The New Palgrave Dictionary of Economics Online. Palgrave Macmillan. 25 March 2008 <http://www.dictionaryofeconomics.com/article?id=pde2008_T000065> doi:10.1057/9780230226203.1710. References can be found in that paper. Selected papers Keynes, J.M. 1939. Professor Tinbergen's method. Economic Journal 49, 558–68. Keynes, J.M. 1940. Comment. Economic Journal 50, 154–6. Tinbergen J., 1939. Statistical Testing of Business Cycle Theories. I: A Method and Its Application to Investment Activity. II: Business Cycles in the United States of America,1919–1932. Geneva: League of Nations. Tinbergen J., 1940. On a method of statistical business cycle research. A reply. Economic Journal 50, 141–54. |
Archives
Octobre 2016
Categories
Tout
|