[Question] Could we solve this email mess if we all moved to paid emails?

Have you ever…

  • Sent an email to some­one in EA and not heard back for many weeks (or more)?

  • Avoided send­ing an email to some­one be­cause you wanted to spare their at­ten­tion, de­spite think­ing there was a fair chance they’d be gen­uinely in­ter­ested?

  • Wanted some way to sig­nal that you ac­tu­ally cared more than usual about this email, but with­out hav­ing to burn so­cial cap­i­tal (such as by say­ing “ur­gent” or “please read”)?

  • Had to ig­nore an email be­cause, even though it might have been in­ter­est­ing, figur­ing that out would sim­ply have been too effort­ful?

I think that 1) prob­lems like these are preva­lent, 2) they have pretty bad con­se­quences, and 3) they could be partly solved by us­ing ser­vices where you can pay to send some­one an email (pay­ment is usu­ally con­di­tional on re­ply).

I’m con­sid­er­ing run­ning a co­or­di­na­tion cam­paign to move the com­mu­nity to us­ing paid emails (in ad­di­tion to their or­di­nary in­box), but be­fore launch­ing that unilat­er­ally I want more con­fi­dence it is a good idea, hence I’m ask­ing this ques­tion.


Email seems bro­ken. This is not that sur­pris­ing: your email is ba­si­cally a to-do list where other peo­ple (and com­pa­nies) can add items for free, with­out ask­ing; and where you’re the only one who can re­move them. We should do some­thing about this.

More broadly, the at­ten­tion econ­omy seems bro­ken. Recog­nis­ing this, many EAs use var­i­ous soft­ware tools to pro­tect them­selves from apps that are en­g­ineered to be ad­dic­tive. This helps at an in­di­vi­d­ual level, but it doesn’t help solve the col­lec­tive ac­tion prob­lem of how to al­lo­cate our at­ten­tion as a com­mu­nity. We should do some­thing about this.

Costly sig­nal­ling and avoid­ing in­for­ma­tion asymmetries

An “in­for­ma­tion asym­me­try” is situ­a­tion where some­one has true in­for­ma­tion which they are un­able to com­mu­ni­cate. For ex­am­ple, sup­pose 10 economists are try­ing to in­fluence gov­ern­ment policy on is­sue X, and one of them ac­tu­ally, re­ally knows what the most effec­tive thing is. Yet, they might not be able to com­mu­ni­cate this to the de­ci­sion-mak­ers, since the re­main­ing 9 have de­grees from equally pres­ti­gious in­sti­tu­tions and ar­gu­ments that sound equally rigor­ous to some­one with­out for­mal train­ing in eco­nomics. In­for­ma­tion asym­me­tries are a key mechanism that gen­er­ate bad equil­ibria.

When it comes to email, this might look as fol­lows: Lots of peo­ple write to se­nior re­searchers ask­ing for feed­back on pa­pers or ideas, yet they’re mostly crack­pots or un­in­ter­est­ing, so most stuff is not worth read­ing. A promis­ing young re­searcher with­out many con­nec­tions would want their feed­back (and the se­nior re­searcher would want to give it!), but it sim­ply takes too much effort to figure out that the pa­per is promis­ing, so it never gets read. In fact, ex­pect­ing this, the ju­nior re­searcher might not even send it in the first place

This could be avoided if peo­ple who gen­uinely be­lieved their stuff was im­por­tant could pay some money as a costly sig­nal of this fact. Ac­tual crack­pots could of course also pay up, but 1) they might be less likely to, and 2) the pay­ment would offset some of the cost of the re­cip­i­ent figur­ing out whether the email is im­por­tant or not.

How the sig­nal­ling prob­lem is cur­rently solved, and why that’s bad

Cur­rently, the sig­nal­ling prob­lem is solved by things like:

  • Spend­ing lots of effort craft­ing in­ter­est­ing-sound­ing in­tros which sig­nal that the thing is worth read­ing, in­stead of just get­ting to the point

  • Burn­ing so­cial cap­i­tal—adding tags like “[Ur­gent]” or “[Im­por­tant]” to the sub­ject line

This is bad, be­cause:

1) It’s a slip­pery slope to a re­ally bad equil­ibrium. I’ve got­ten emails with ti­tles like “Ja­cob, is ev­ery­thing alright be­tween us?” be­cause I didn’t e.g. buy a wa­ter bot­tle from some com­pany. This is what we should ex­pect when com­pa­nies fight for my at­ten­tion with­out any way to just di­rectly pay for it. Even within the EA com­mu­nity, if our only way of al­lo­cat­ing im­por­tance is by draw­ing upon very se­ri­ous vo­cab­u­lary, we’ll cre­ate an in­cen­tive for ex­ag­ger­a­tion, differ­en­tially favour­ing those less scrupu­lous about this prac­tice, and chip away at our abil­ity to use shared-cues-of-im­por­tance when it re­ally mat­ters.

2) The main thing pro­tect­ing us from this in­side a smaller com­mu­nity is that peo­ple want to pre­serve their rep­u­ta­tions. But if you’re un­sure how im­por­tant your thing is, and mis­la­bel­ing it means po­ten­tially cry­ing-wolf and risk­ing your rep­u­ta­tion, this usu­ally makes it more worth it to just avoid the tag. Which means that we lose out on all those times when your thing ac­tu­ally was im­por­tant and us­ing the tag would have com­mu­ni­cated that.

3) It puts the re­cip­i­ent be­tween a rock and a hard place, and they’re not be­ing com­pen­sated for it. If you mark some­thing as “[Ur­gent]” that ac­tu­ally is ur­gent, and the per­son re­sponds and does what you want, you’ve still pre­sented them with the choice be­tween sac­ri­fic­ing some abil­ity to freely pri­ori­tise their tasks, and sac­ri­fic­ing some part of the qual­ity of your re­la­tion­ship. There should be some easy way for you to com­pen­sate them for that.

4) It’s way too coarse-grained. There’s not re­ally any way of say­ing:

“This is kinda im­por­tant, but not that ur­gent, though it would prob­a­bly be good if you read it at some point, though that de­pends on what else is on your plate”

apart from writ­ing ex­actly that—but then you’re mak­ing a com­pli­cated cog­ni­tive de­mand, which has already burnt lots of at­ten­tion for the re­cip­i­ent.

Brief FAQ

What if re­plac­ing email with paid emails puts us in an­other equil­ibrium that’s bad for un­ex­pected rea­sons?

At the mo­ment, it doesn’t seem fea­si­ble for us to use this to re­place email. There isn’t even soft­ware available for do­ing that com­pletely. Rather, peo­ple would con­sent to re­ceiv­ing paid mes­sages (for ex­am­ple via earn.com, see be­low) in ad­di­tion to hav­ing their reg­u­lar in­box.

What if peo­ple don’t have enough money?

As men­tioned above, send­ing stan­dard emails are still an op­tion. Yet this be­comes a prob­lem in the world where we move to the equil­ibrium where a stan­dard email is taken to sig­nal “I didn’t pay for this, so it’s not that im­por­tant”. Then I can imag­ine grants for “email costs” be­ing a thing, or that the benefits of the new equil­ibrium out­weigh this cost, or that they don’t. I’m un­cer­tain.

Wouldn’t this waste a lot of money?

Not re­ally, as­sum­ing that the peo­ple who you send money to are at least as effec­tive at spend­ing it as you are, which seems likely if this gets used within the EA com­mu­nity.

If this is ba­si­cally right: then what do we do?

Earn.com is a site which offers paid emails. For ex­am­ple, you can pay to mes­sage me at earn.com/​ja­cob­ja­cob/​

If this seems like some­thing that could solve the cur­rent email mess, we should co­or­di­nate to get a crit­i­cal mass of the com­mu­nity to sign-up, and make their pro­file url:s available. (Com­pare this to how we’ve pre­vi­ously started us­ing things re­ciproc­ity.io and Cal­endly.)

I’d be happy to co­or­di­nate such a cam­paign, but I don’t want to do it un­til I’m more con­fi­dent it would be a good thing. I hope to use this ques­tions to help figure that out.

(For the record, I have no re­la­tion to earn.com and would not benefit per­son­ally by oth­ers join­ing, be­yond the ob­vi­ous pos­i­tive effects on the com­mu­nity. They sim­ply seem like the best available op­tion for do­ing this. They have a pretty solid team, and are used by some very se­nior VCs like Marc An­dreessen and Keith Rabois.)