Consider the standard Diamonds coconut economy. One could imagine that people who search longer, have a different likelihood of matching someone to trade with.

Has there been something done akin this (in any part of matching theory)?

Second Example Say, we want to match liver transplants between donors $d$ and recipients $r$. The longer anyone waits, the higher the likelihood of being matched with someone from the other side.

Hence, we would have a distribution of $r$ and $d$ w.r.t the time that they have already been waiting. And instead of an aggregate matching function $M(r, d)$ we would have something that takes into account these distributions.

What has been done in this area so far?

  • $\begingroup$ Your search query for google might be 'dynamic matching'. There seems to be some work done on that. I happen to wonder right now whether the relationship between dynamic matching markets and dynamic price mechanisms have been contrasted. $\endgroup$
    – ramazan
    Sep 21, 2015 at 7:03
  • 1
    $\begingroup$ My impression is that this is already taken into account into the traditional search setup. The issue is that per-unit-of-time waited the likelihood of getting matched is the same. There is no list that you move up in as you wait longer. If you search many days in a row, you are more likely to end up matched, but at any point in time you are as likely to be matched as anyone else looking for work at that point. $\endgroup$
    – Fix.B.
    Mar 31, 2016 at 3:10

1 Answer 1


The paper is a couple of years old by now, but I believe the first people to look seriously at the issues you're thinking of are Akbarpour, et al.

It might be useful to look at that paper and the handful of others that cite it. The paper was originally called 'Dynamic Matching Market Design', so it may have more citations under that name.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.