I am struggling understanding the concept of rational expectations. Can someone explain in simple words the concept of rational expectations? Does someone know a good (online) source explaining rational expectations?

  • $\begingroup$ Just know that it is a terrible assumption. Economists have shown in experiments over and over and over that RE does not hold. People deviate systematically from rationality. The representative agent assumption is equally stupid. $\endgroup$ – 123 Sep 7 '18 at 14:53
  • 4
    $\begingroup$ Ok, this guy 123 is wrong. Assuming rationality is one thing, modeling agents in macroeconomic models with rational expectations is an entirely different thing. Cochrane's take on the matter is a good read. $\endgroup$ – Pedro Cavalcante Sep 8 '18 at 20:55
  • 1
    $\begingroup$ Pedro: Very interesting link. Thanks for pointing it out. $\endgroup$ – mark leeds Sep 9 '18 at 1:23
  • 1
    $\begingroup$ Rational Expectations isn't assuming people correctly predict the future, just that they don't see what happened today, what they thought would happen tomorrow and change expectations adaptively. People use all sorts of information during expectations formations - that's what Rational Expectations is about, not that people have superpowers. $\endgroup$ – Pedro Cavalcante Sep 9 '18 at 18:26
  • 1
    $\begingroup$ And I'm not even mentioning the econometric implications of RE. You are wrong, your entire conception of RE is basically a strawman. Not only that, you're also confusing what assuming rationality in a microeconomic modelling context with Rational Expectations. $\endgroup$ – Pedro Cavalcante Sep 9 '18 at 18:28

The concept of rational expectations can indeed be confusing.

One of the nicest pieces that I read online on rational expectations is the following:


It very nicely describes the concept of rational expectations in very simple words.

The most important thing one needs to understand is that rational expectations do not imply that all people act optimal under all circumstances, it rather implies that people behave optimally given the information they have.

  • $\begingroup$ Hi Gio: It is a clear article. But, to be honest, from an econometric perspective, it is mis-leading because it says that, in RE, economic agents can have limited information or whatever information you assume them to have. This is not really true. Suppose one has a model that says tomato prices are AR(1). Then RE assumes that the farmer of tomatoes also believes that tomatoe prices are AR(1). That might be viewed as limited but for the article to say that the information of the economic agent can be limited could be mis-leading since he has the same info as the modeler. Thanks for link. $\endgroup$ – mark leeds Sep 10 '18 at 19:37

The concept of rational expectations asserts that outcomes do not differ systematically (i.e., regularly or predictably) from what people expected them to be. The concept is motivated by the same thinking that led Abraham Lincoln to assert, “You can fool some of the people all of the time, and all of the people some of the time, but you cannot fool all of the people all of the time.” From the viewpoint of the rational expectations doctrine, Lincoln’s statement gets things right. It does not deny that people often make forecasting errors, but it does suggest that errors will not persistently occur on one side or the other.

from Tomas Sargent (https://www.econlib.org/library/Enc/RationalExpectations.html)

  • $\begingroup$ Hi: I've read a lot about RE and my belief is that one can only understand it with respect to its econometric implications. There are so many discussions but possibly the best one I've seen is 171-179 of the paper "on econometric models with rational expecations" by broze and szafarz. of course, you can read the whole thing but those first 8 pages are really good. If you can't obtain it, let me know and I can scan it and send it to your email address. Pesaran's text is the best full book discussion of it that I know of. – mark leeds 2 hours ago $\endgroup$ – mark leeds Sep 7 '18 at 21:24

Rational Expectations - as opposed to Adaptive Expectations - is a theoretical framework in which agents form expectations not solely based on past experiences but also on current information, priors and what they think is going to happen.

It doesn't imply perfect prediction of the future, just that a policymaker can't systematically fool agents, at the risk of losing credibility. This was a big leap forward in terms of macroeconomic theory in the 70s.

EDIT: Mark's comment reminded me of an important aspect of RE. Before it, models assumed agents didn't know the model. Part of the Lucas Critique is that policymakers can't statically estimate parameters and think that agents aren't responsive to policy trying to explore trade-offs and induce certain states.

  • 1
    $\begingroup$ Pedro: There are so many definitions so I'm not saying you are wrong but, I'll give a different definition here. From an econometric perspective, RE says that the agent ( say the farmer being modelled in a wheat demand model ) has the same model as the econometrician so his expectations are the same as those in the model itself. It's a really interesting viewpoint ( it's origins are in the Muth, 1961 paper ) and some say too strong of an assumption. $\endgroup$ – mark leeds Sep 8 '18 at 21:04
  • $\begingroup$ I've just added this (very relevant) point. Thanks, Mark. $\endgroup$ – Pedro Cavalcante Sep 8 '18 at 23:41
  • $\begingroup$ Pedro: your welcome and good to meet you. $\endgroup$ – mark leeds Sep 9 '18 at 1:30

I will provide an actual answer because I strongly believe that embedding the rational expectations (RE) assumption into macroeconomic models is ridiculous.

I will try to crystalize this in a concise way. Anytime we try to use a macroeconomic model to forecast outcomes we are actually creating a model that incorporates, via probability distribution, all possible outcomes. RE assumes that nothing unforeseen occurs (it is whenever nothing unforeseen occurs that rational agents can act optimally) and that rational agents are then able to compute profit maximizing choice paths. Further, any behavioral deviations necessarily produce sub-optimal paths for choice variables. This is akin to assuming that all economic actors know the model governing the DGP within which they operate. And this yields (2): agents can only form RE whenever they know the true specification of the economic model governing their world and its exact parameterization. You decide for yourself if you think these are good assumptions.

This yields an extreme (and perhaps dangerous) system of thinking; the supposition that economic models ought to consistently provide reliable quantitative predictions about future outcomes, for example. It is also ironic that modern macroeconomists micro-found current models to attain quantitative predictions over an ordinal outcome.

See the work of Roman Frydman. A good starting point is "Towards an Understanding of Market Processes: Individual Expectations, Learning, and Convergence to Rational Expectations Equilibrium". AER, 1982.

You might also look into the behavioral and experimental macroeconomics literature.

edit: I will add -- Evans, G. W. and G. Ramey (2006) Adaptive Expectations, Under parameterization and the Lucas Critique. Journal of Monetary Economics for assertion (2)

edit2: My memory was incorrect and erik (below in comments) is correct. Frydman didn't win the NP. It was his co-founder of his institution at Columbia university Edmund Phelps.

  • $\begingroup$ RE isn't assuming "nothing unforeseen occurs". Good god, read the Lucas Critique. You're attacking a strawman. $\endgroup$ – Pedro Cavalcante Sep 9 '18 at 18:59
  • $\begingroup$ I have definitely read that work. You are obviously the one who is out of touch with literature. You come across as someone without a PhD who was browbeaten with RE and is now a full-on disciple regurgitating the arguments of others. And you keep saying straw man as though you've provided a more accurate definition of RE than the one I take from Frydman, Heckman, and others. In fact, you've provided a definition of RE where RE gets to be anything other than fully-adaptive behavior. $\endgroup$ – 123 Sep 9 '18 at 19:03
  • $\begingroup$ This isn't my area (I work with education and labor), but it doesn't take much to realise you're attacking a strawman. Please, name one economist who thinks "nothing unforseen occurs". $\endgroup$ – Pedro Cavalcante Sep 9 '18 at 19:05
  • $\begingroup$ P.S.: The funny thing I don't even agree with RE. I much rather prefer Gabaix's approach of bounded rationality and intertemporal myopia, which uses many insights from behavioral econ. $\endgroup$ – Pedro Cavalcante Sep 9 '18 at 19:10
  • 1
    $\begingroup$ Also, if you either of you know why the authors of econometric RE publications write in such a way that only the crew of mcallum, taylor, sims, pesaran,, sargent, wallis, hansen, wallace, lucas and a few others can follow them, I'd love to know. Wouldn't they want to spread the word by making their work more readable ? Often nothing is even defined ? Somehow I was able to understand Lucas paper but that was a miracle and maybe specific to Lucas because I haven't tried to read a lot of his material. Maybe a lot of his material is readable ? $\endgroup$ – mark leeds Sep 10 '18 at 4:32

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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