Monday, October 11, 2010

2010 Nobel prize in economics for unemployment!

This year’s Nobel prize craze finally ended with the announcement of the prize in economics to Peter Diamond, Dale Mortensen and Christopher Pissarides for “their analysis of markets with search frictions”. It is a prize for (structural) unemployment! I have heard and read some of Diamond’s stuff but not other’s. Here is an interview with Peter Diamond.

This blog post is largely derived from an extended note by the prize committee. Just to help me understand what’s up with the DMP model. It is pretty fascinating. Even though the main conclusion of their work looks obvious, they modeled it and systematically applied it in the real world. Their work has a Keynesian twist and slaps the classicists notion that markets clear itself. Tyler Cowen thinks of Pissarides “as the least Keynesian of the trio.” Here is Cowen on Diamond and Mortensen. Krugman is happy that the trio won the prize.

Diamond, Mortensen and Pissarides showed that an unregulated market does not clear by itself, as has been asserted by classical economics. The classical economists believe that markets always clear because prices are determined in such a way that demand matches supply as there are no transaction costs and there is perfect information. In reality, it ain’t so. Buyers and sellers face costs in their attempts to locate each other (“search”) and meet pairwise when they come into contact (“matching”). They explored how price formation works in a market with search frictions. How much price dispersion-- if the law of one price should be expected to hold in markets with frictions-- will be observed and how large are the deviations from competitive pricing?

Buyers are not able to find sellers, or vice versa, all the time. Even if they did, there might be disagreements in price of the product in question. This means that both will continue searching for the perfect deal. This process of finding the desired outcome is not frictionless. The trio clarified this by both using economic models and applying that to real life. They analyzed how (frictional) unemployment works and persists in an economy. Agents in an economy always look for cooperative patterns in order to meet their desired goal. For instance, in order to reach a settlement, there is always a search for cooperative deal between a buyer and a seller of a product, between employers and prospective employees, and between firms and their suppliers. When a producer produces an item, it does not sell automatically, unless it is pre-ordered, which usually does not happen all the time. This means that the seller will have to wait until he finds the right customer willing to pay the right price, and vice versa.

In 1971, Peter Diamond examined how prices are formed on a market where buyers look for the best possible price and sellers simultaneously set their best price while taking buyers’ search behavior into account. A small search cost would drastically change the outcome, which is not factored into the classical labor market model. He showed that the mere presence of costly search and matching frictions does not suffice to generate equilibrium price dispersion. He found that even a small search cost moves the equilibrium price away from the competitive price (typically exists under perfect competition). The only equilibrium outcome is the monopoly price. This is dubbed the “Diamond paradox”. This means that a small search friction can have a large effect on price outcomes, and it would not lead to any price dispersion at all.

Mortensen and Pissarides extended Diamond’s model to this notion to labor markets. Their work sheds light on how economic policy and regulation can affect unemployment, job vacancies, and wages. It relates to figuring out unemployment insurance or hiring and firing rules. Their work shows that a more generous unemployment benefits give rise to higher unemployment and longer search times.

Their research showed that an unregulated search market does not give rise to an efficient outcome. Aggregate welfare is not necessarily higher with more search since search itself is costly. Resource utilization could be either low or high as the search and matching process involves costs. There are external effects that an agent does not know of. For instance, intense search for jobs by an agent means more difficulty for other job seekers to find employment because it will be easy for companies to find desired personnel, and vice versa. Additionally, unlike in the unregulated classical models, when search costs exist, there could be several optimal points, but only one superior optimal point. This is where the role of the government comes in. It can find ways to ensure that the search process results in a superior outcome when the final deal is sealed. Unutilized resources can be put to good use by intervention from an external agent.

The search and matching environment can lead to macroeconomic unemployment problems as coordinating trade does not match one-to-one. This provides a rationale for “aggregate demand management” to steer the economy towards the best equilibrium. Diamond’s work is viewed as “a careful analysis, using microeconomic foundations, to analyze some of the central themes of Keynes’s business-cycle theory.” Coordination problems feature on both of their writings. An appropriate intervention by the government can address coordination problems.

Their work on the determinants of unemployment (DMP model) can be used to explain the position of the Beveridge curve and the location of the economy on the curve. A Beveridge curve shows that the labor market fluctuates between situations of either high unemployment and few vacancies or low unemployment and many vacancies. If unemployment increases when vacancies go down, then it is related to the demand for labor and is related to business cycle. But, if vacancies go up and unemployment also increases, then the labor market is not performing well, signaling that there could be weaker matching efficiency resulting in long-term unemployment. Understanding why this happens is important to reducing unemployment.

The DMP model addresses the following questions:

    1. How workers and firms jointly decide whether to match or to keep searching
    2. In case of a continued match, how the benefits from the match are split into a wage for the worker and a profit for the firm
    3. Firm entry, i.e., firms’ decision to “create jobs”
    4. How the match of a worker and a firm might develop over time, possibly leading to agreed-upon separation.

The model is used to examine the effects of policies concerning hiring costs, firing costs, minimum wage laws, taxes, and unemployment benefits on unemployment and economic welfare.It also used to analyze how aggregate shocks are transmitted to the labor market and lead to cyclical fluctuations in unemployment, vacancies, and employment flows.

As a side note, here is an abstract from a paper that uses search theory in marriage where older men appear to be attractive in the marriage market for younger women!

It is commonly observed that across societies and time, women tend to marry older men. The traditional explanation for this phenomenon is that wages increase with age and hence older men are more attractive in the marriage market. The explanation holds even where differences in fertility between men and women are taken in account. This explanation, however, involves an implicit assumption about female specialization in home production - an assumption that does not generally hold, especially in modern times. This paper shows that a marriage market equilibrium where women marry earlier in life than men can be achieved without making any assumptions about the wage process or gender roles. The only driving force in this two sided search model is the asymmetry in fertility horizons between men and women. When the model is calibrated with Census Data, the average age at first marriage and the pattern of the sex ratio of single men to single women over different age groups mimics the patterns observed in developed countries during the last decade (e.g. France, the U.S. and Sweden). However, the fit is less accurate for developing countries and for earlier decades in developed countries. This result may indicate a more important role of social norms and wages in the determination of marriage pattern in those cases.