Against boots theory

Link post

The rea­son that the rich were so rich, Vimes rea­soned, was be­cause they man­aged to spend less money.

Take boots, for ex­am­ple. He earned thirty-eight dol­lars a month plus al­lowances. A re­ally good pair of leather boots cost fifty dol­lars. But an af­ford­able pair of boots, which were sort of OK for a sea­son or two and then leaked like hell when the card­board gave out, cost about ten dol­lars. Those were the kind of boots Vimes always bought, and wore un­til the soles were so thin that he could tell where he was in Ankh-Mor­pork on a foggy night by the feel of the cob­bles.

But the thing was that good boots lasted for years and years. A man who could af­ford fifty dol­lars had a pair of boots that’d still be keep­ing his feet dry in ten years’ time, while the poor man who could only af­ford cheap boots would have spent a hun­dred dol­lars on boots in the same time and would still have wet feet.

This was the Cap­tain Sa­muel Vimes ‘Boots’ the­ory of so­cioe­co­nomic un­fair­ness.

– Terry Pratch­ett, Men at Arms

This is a com­pel­ling nar­ra­tive. And I do be­lieve there’s some truth to it. I could be­lieve that if you always buy the cheap­est boots you can find, you’ll spend more money than if you bought some­thing more ex­pen­sive and re­li­able. Similar for lap­tops, smart­phones, cars. Espe­cially (as Siderea notes, among other things) if you know how to buy ex­pen­sive things that are more re­li­able.

But it’s pre­sented as “the rea­son that the rich [are] so rich”. Is that true? I mean, no, ob­vi­ously not. If your pre-tax in­come is less than the amount I put into my sav­ings ac­count, then no amount of “spend­ing less money on things” is go­ing to bring you to my level.

Is it even a con­tribut­ing fac­tor? Is part of the rea­son why the rich are so rich, that they man­age to spend less money? Do the rich in fact spend less money than the poor?

That’s less ob­vi­ous, but I pre­dict not. I pre­dict that the rich spend more than the poor in to­tal, but also on boots, lap­tops, smart­phones, cars, and most other things. There might be ex­cep­tions where rich peo­ple con­sume less of the thing than poor peo­ple—bus tick­ets, for ex­am­ple—but I think if you group spend­ing in fairly nat­u­ral ways, the rich will spend more than the poor in al­most ev­ery group.

  • Maybe they spend less money on their daily wear boots, but own more pairs of shoes for differ­ent oc­ca­sions. Or maybe they de­cide that they care about other things than life­time cost for their daily wear boots, and spend more on those, too. (Be­ing rich means they can af­ford to care about other things than life­time cost.)

  • Ap­par­ently fa­mous peo­ple of­ten get comped meals, but I bet most of them still spend more money on food than I do.

  • I spent £500 on a lap­top in 2013, and be­fore that, £300 in 2008. If I’d gone for £200 lap­tops each time, maybe they would only have lasted two years each. But if I weren’t a techno-masochist, maybe I’d re­al­ize that us­ing old lap­tops ac­tu­ally kind of sucks, and I’d up­grade far more of­ten. My work lap­top, bought by peo­ple who want me to be max­i­mally effec­tive at my job, cost over £1000 and isn’t go­ing to last ten years.

  • Fi­nan­cial ser­vices are a case where I’d guess the rich and the poor spend money on very differ­ent things. I as­sume the rich don’t have to pay to cash a cheque, and very rarely visit loan sharks. But the poor rarely have Amex Plat­inum cards ($550/​year), or per­sonal ac­coun­tants. (Maybe it’s un­fair to count those be­cause they save you money in other ar­eas?)

  • Buy­ing a house may be cheaper in the long run than rent­ing a similar house nearby. But rich peo­ple tend to live in nicer houses and/​or nicer ar­eas.

Those are all guesses. I don’t have good data on this, and I’d love to see it if you do.

For what data I do have, the first google re­sult was this page from the UK’s Office of Na­tional Statis­tics. Speci­fi­cally, look at figure 4, “In­dexed house­hold in­come, to­tal spend­ing and spend­ing by com­po­nent by in­come decile, UK, FYE 2019”.

They split house­holds into ten in­come lev­els, and look at four cat­e­gories of spend­ing plus to­tal spend­ing. Each of those is a near-strictly in­creas­ing line from “poor peo­ple spend less” to “rich peo­ple spend more”. (I see two blips: the 90th per­centile of in­come spends slightly less on hous­ing than the 80th, and the 70th spends slightly less on food and non-al­co­holid drinks than the 60th. The other cat­e­gories are trans­port, and recre­ation and cul­ture. Th­ese four are the largest spend­ing cat­e­gories on av­er­age across all in­come lev­els. The graph also has dis­pos­able in­come, which I think is ir­rele­vant for cur­rent pur­poses.)

(I re­peat that this spe­cific data is not strong ev­i­dence. The source for it is the liv­ing costs and food sur­vey, which might have more de­tail. (Link goes to the pre­vi­ous year’s ver­sion be­cause that’s what I could find.) Un­for­tu­nately it’s not open ac­cess. It might be freely available if I reg­ister, but I don’t care enough to try right now. In any case, we’d also want data from out­side the UK.)

There will ob­vi­ously be some ex­cep­tions. There will be some rich peo­ple who spend less money than some poor peo­ple. There will prob­a­bly even be some rich peo­ple who spend less money than some poor peo­ple, and would not be rich oth­er­wise. But as a gen­eral the­ory for why the rich are rich? I just don’t buy it.

I be­lieve boots the­ory points to­wards one com­po­nent of so­cioe­co­nomic un­fair­ness. But boots the­ory it­self is sup­posed to be a the­ory of why the rich are so rich. It’s very clear about that. It’s clearly wrong, and I pre­dict that even a weak­ened ver­sion of it is wrong.


To be a lit­tle more pre­cise, I think boots the­ory as writ­ten makes three in­creas­ingly strong claims, that we could think of as “lev­els of boots the­ory”:

  1. Be­ing rich en­ables you to spend less money on things. (More gen­er­ally: hav­ing in­cre­men­tally more cap­i­tal lets you spend in­cre­men­tally less money. Also, be­ing rich is su­per con­ve­nient in many ways.) This phe­nomenon is also called a ghetto tax.

  2. Also, rich peo­ple do in fact spend less money on things.

  3. Also, this is why rich peo­ple are rich.

All of these lev­els have stronger and weaker forms. But I think a quick look at the world tells us that the first level is ob­vi­ously true un­der any rea­son­able in­ter­pre­ta­tion, and the third level is ob­vi­ously false un­der any rea­son­able in­ter­pre­ta­tion. The sec­ond I pre­dict is “ba­si­cally just false un­der most rea­son­able in­ter­pre­ta­tions”, but it’s less ob­vi­ous and more de­pen­dent on de­tails. There may well be weak forms of it that are true.

It may be that most peo­ple, when they think of boots the­ory, think only of lev­els one or two, not level three. I don’t know if you can read this quora thread that I found on Google. It asks “How ap­pli­ca­ble to real life is the Sam Vimes “Boots” The­ory of Eco­nomic In­jus­tice?” The an­swers mostly agree it’s very ap­pli­ca­ble, but I think most of them are on level one or two. (The one talk­ing about lev­er­age seems like level three, if it’s talk­ing about boots the­ory at all. I’m not con­vinced it is.)

But it seems to me that boots the­ory is usu­ally pre­sented in whole in its origi­nal form. Its origi­nal form is suc­cinct and well writ­ten. When peo­ple want to com­ment on it, they very of­ten in­clude the very same quote as I did. And the origi­nal form starts by very clearly tel­ling us “this is a the­ory of why the rich are so rich”. It is very ob­vi­ously level three, which is very ob­vi­ously wrong.

So I have a few com­plaints here.

One is, I get the im­pres­sion that most peo­ple don’t even no­tice this. They link or quote some­thing that starts out by say­ing very clearly “this is a the­ory of why the rich are so rich”, and they don’t no­tice that it’s a the­ory of why the rich are so rich.

(I wouldn’t be too sur­prised (though this is not a pre­dic­tion) if even the au­thor didn’t no­tice this. Maybe if you had asked him, Terry Pratch­ett would have said that no, ob­vi­ously Sam Vimes does not think this is why the rich are so rich, Sam Vimes just thinks this is a good illus­tra­tion of why it’s nice to be rich.)

This dis­con­nect be­tween what a thing ac­tu­ally says, and what peo­ple seem to think it says, just both­ers me. I feel the de­sire to point it out.

Another is, I think there’s a motte-and-bailey go­ing on be­tween lev­els one and two. A quora com­menter says it’s “far more ex­pen­sive to be poor than it is to be rich, both in a per­centage of in­come re­spect and a di­rect effect”. He gives ex­am­ples of things that rich peo­ple can spend less money on, if they choose. He doesn’t provide data that rich peo­ple do spend less money on these things. Another de­scribes how be­ing rich lets you save money on food sta­ples by stock­ing up when there’s a sale. He doesn’t provide data that rich peo­ple do spend less money on food or even sta­ples. You could cer­tainly make the case that nei­ther of these peo­ple is ex­plic­itly claiming level two. But I do think they’re hint­ing in that di­rec­tion, even if it’s not de­liber­ate.

And re­lat­edly: if we want to help peo­ple es­cape poverty, we need to know on what lev­els boots the­ory is true or false.1 If we want to know that, we need to be able to dis­t­in­guish the lev­els. If “boots the­ory” can re­fer to any of these lev­els, then sim­ply call­ing boots the­ory “true” (or even “false”) is un­in­for­ma­tive. We need to be more pre­cise than that. To be fair, the quora com­menters make spe­cific falsifi­able claims, which is com­mend­able. But the claims are meant to be spe­cific ex­am­ples of a gen­eral phe­nomenon, and the gen­eral phe­nomenon is sim­ply “boots the­ory”, and it’s not clear what they think that means.

I ad­vise that if you talk about boots the­ory, you make it clear which level you’re talk­ing about. But maybe don’t use that name at all. If you’re talk­ing about level one, the name “ghetto tax” seems fine. If you do want to talk about lev­els two or three, I don’t have a good al­tier­na­tive name to sug­gest. But since I don’t think those lev­els are true, I’m not sure that’s a big prob­lem.

  1. I’m not too con­fi­dent about this, and I don’t want to get too dis­tracted with ob­ject-level claims about how to ac­tu­ally fight poverty. But my sense is that: to the ex­tent that level two is true, giv­ing some­one money fairly re­li­ably sets up pos­i­tive feed­back loops that help them save more money in fu­ture. To the ex­tent that it’s not true, these feed­back loops don’t come for free. Maybe we can seek out spend­ing cat­e­gories where it is true, or groups of peo­ple for whom it is true. Maybe we can teach peo­ple how to find and take ad­van­tage of these feed­back loops. If even level one isn’t true, we don’t get these loops at all. Of course, maybe it’s worth giv­ing peo­ple money even if we don’t get the feed­back loops.