@dglid Metaculus says:

NBER may declare that a recession has occurred months after the fact, including after popular media consensus. If NBER does not declare that a recession has occurred in the relevant time period within 2 years, the question will resolve as No. For example, for the period ending in January 1, 2023, the question will resolve (at the latest) in January 1, 2025

Does this mean the market close date should be extended until Jan 1st 2025?

I'd rather it not, but if the intent is to exactly mirror metaculus, then I suppose it should be reopened and extended.

@chrisjbillington The resolution is Jan 1 2025 (or when NBER declares a recession, which may come earlier).

The question has always been intended to exactly mirror metaculus. The 2 year period is important. NBER takes a while to determine recessions.

As for close vs open, I reopened because as discussed on previous markets like this, it's generally not necessary follow the Metaculus close date exactly - they use a different scoring system and so they can't change their close date, we can without problems. (This is purely talking about close vs open, not about resolution date)

@jack Yes, when I said "I'd rather it not", I was griping about the lengthy delay before resolution, it should definitely be open if it's not resolving yet.

@chrisjbillington I approve. Thanks for extending @jack 👍

This market closes in a few days, but the metaculus question closes Jan 1st 2024 and resolves Jan 1st 2025 (If I'm reading right). Are we following them, or actually closing or resolving much sooner?

@chrisjbillington Updated to EOY 2023. It’s possible the close date changed on Metaculus after this market was originally created.

This is a ludicrous arbitrage opportunity!!! This is the question with the widest gap between Manifold and Metaculus (Currently at 20%), with tons of predictors on both sites. Everyone here go sign up for Metaculus and lets pick up some free internet points together.

Metaculus has a 61% prediction on a recession in the 'period ending before 2024'?

And why even arbitrage between manifold and metaculus?

@jacksonpolack So that when one of the two platforms shuts down in embarrassment for predicting this question so badly, you'll be raking it in on the surviving platform. (Oh wait, that's the opposite of arbitrage...)

Metaculus has two questions there, one for 2023 and one for 2024.

Manifold also has ~50% on recession this year. https://manifold.markets/VivaLaPanda/will-the-us-officially-enter-a-rece

@chrisjbillington got any insider info?

@kenakofer No insider info. My strong bets are based on some commentators I trust on the matter (e.g. https://www.nytimes.com/2022/07/26/opinion/recession-gdp-economy-nber.html) saying that the NBER effectively bases recession declarations almost exclusively on employment, with GDP contraction being a useful proxy but practically irrelevant if it conflicts with employment data. I have seen I think not one claim from a reputable economist that they expect the NBER to declare a recession for 2022. I haven't gone looking for such claims, mind you, but neither have I gone looking for claims they wont declare a recession, and I have seen a bunch of them. Interested if anyone has links to reputable people arguing to the contrary, though.

From what I can see so far, the expectation that there will be a recession declared seems to be entirely based on the treatment of 2 quarters of negative GDP growth as a strong indicator, but this expectation seems to be held exclusively by those who haven't done their homework on the matter or otherwise seem untrustworthy. There are plenty of commentators who seem to be reflexive contrarians chiming in who can be safely ignored - online economics discussions seem to be rife with these sorts. You know the type - the kind who complains that they changed the CPI basket of goods to suppress the "true" inflation rate, or that they changed the definition of a recession to suppress that the US was in one in 2022. If someone sounds like they have a chip on their shoulder about the economic establishment or the Fed, opinion goes in bin (even if their main gripe is legitimate - defensiveness about it infects the rest of a person's thinking far too often).

Anyway, yeah, so I think it's about employment. The best/most well-known metric of (un)employment designed to be used as a recession indicator shows not even a hint of one in 2022 (the SAHM indicator is treated as indicating a recession if it rises above 0.5):

https://fred.stlouisfed.org/series/SAHMCURRENT

The Sahm rule was created in order to be a real-time recession indicator, as would be needed for automatic stabilisation mechanisms as advocated for by the creator, Claudia Sahm. Twitter thread on the rationale here: https://twitter.com/Claudia_Sahm/status/1129195112733523968

I went hard on the below market too, since up until March unemployment data the Sahm rule was not showing recession yet:

/OliverS/will-there-be-a-recession-in-the-us

If anyone thinks I'm very wrong on this, I'd be eager to hear it, but yeah basically I'm betting as if the Sahm indicator might as well be the definition of a recession.

This is almost the same question as /jack/did-the-us-officially-enter-a-reces , but allowing NBER 2 years to make the determination vs 1 year.

I don't know the first thing about how recessions are decided, but Metaculus is at 28% right now. These two markets are going to resolve the same way!