The Collapse of a Medical Labor Clearinghouse (and Why Such Failures Are Rare)

Abstract

The collapse of the clearinghouse for the entry-level gastroenterology labor market offers a unique opportunity to study how stable clearinghouses succeed and fail. To explore the reasons for the failure of the clearinghouse (and why failures of this kind of clearinghouse have been so rare), we conduct an experimental investigation of demand shocks of the kind that occurred in the gastroenterology market. We find that a reduction in demand for positions leads to the collapse of the match only when it is detectable by firms before being detected by workers (as in the unexpected shock that took place in 1996, which could be seen by firms in their reduced applicant pools). Simple demand and supply imbalances do not seem to interfere with the operation of the centralized match. Our results suggest an affirmative answer to the question posed by market participants about whether the clearinghouse could be successfully restarted, and that this would relieve some of the distress now reported in that market, by allowing it to operate later, at a more uniform time, and with more national scope.

School choice: an experimental study

“…While we are not aware of any previous experimental studies of school choice mechanisms, there have been experimental studies of other one-sided and two-sided matching problems motivated by various real world applications. These include Olson and Porter (1994), Nalbantian and Schotter (1995), Harrison and McCabe (1996), Kagel and Roth (2000), Ünver (2001), Haruvy, Roth andÜnver (2001), Chen and Sönmez (2002), and McKinney, Niederle and Roth (2004). Our experiment differs from the above experiments in both the particular mechanisms compared and the potential field applications of the mechanisms.…”

Competitive Wages in a Match with Ordered Contracts

“…Niederle and Roth (2003b, 2004 and Niederle, Proctor and show that the labor market for gastroenterology fellows, after they stopped using a centralized match, once more unraveled, with thin and dispersed markets, and reduced mobility. 15 It seems that the market 13 Peter Coles (2005) discusses the extent of manipulation of strategies in simple matching environments. On the other hand, Fuhito Kojima and Parag A. Pathak (2006) show that as markets are large and but rank order lists are short, there is not much room for manipulation.…”

“…14 Georgy Artemov (2006) points out that noise in standard matching models with mutliple contracts may have large effects on wages. 15 The failure of the gastroenterology fellowship match is one of the rare instances in which a match which produces stable outcomes has been abandonned. C. Nicholas McKinney, Niederle and Roth (2005) argue that for gastroenterology fellows is one in which programs scramble to deal with congestion issues, in which programs make exploding offers, and start making offers earlier and earlier (Niederle, Proctor and Roth 2006).…”

What Have We Learned From Market Design?

“…20 The market for gastroenterology fellows was organized via a stable labor market clearinghouse (a "match") from 1986 through the late 1990's, after which the match was abandoned (following an unexpected shock to the supply and demand for positions in 1996; see Mckinney, Niederle and Roth, 2005). This provided an opportunity to observe the unraveling of a market as it took place.…”