Skip to content

Accounting for variation and uncertainty

[cat picture]

Yesterday I gave a list of the questions they’re asking me when I speak at the Journal of Accounting Research Conference.

All kidding aside, I think that a conference of accountants is the perfect setting for a discussion of of research integrity, as accounting is all about setting up institutions to enable trust. The challenge is that traditional accounting is deterministic: there’s a ledger and that’s that.

In statistics, we talk all the time about accounting for variation and uncertainty. Maybe “accounting” is more than a metaphor here, and maybe there’s more of a connection to the traditional practices of accounting than I’d thought.


  1. Jonathan (another one) says:

    “there’s a ledger and that’s that”

    As atonement for my accounting joke of yesterday, that’s not right. Lots of accounting issues involve probabilistic projections. Probably the biggest one is reserve allowances. A big company has been sued, and it knows it’s probably going to settle, or maybe it has a substantial chance of losing. Accounting theory, in order to fairly reflect the state of the firm, requires the firm to book an expectation of loss today in a reserve account. So, right from the start, you have to model the expected loss.

    But that’s when it gets really interesting. (Accounting! Interesting, the economist concedes.) Management may want a cookie jar of stashed away earnings to massage their numbers. If I have a good quarter, I might have an incentive to overstate my reserves for some future contingency, conveniently reversing myself in a bad quarter in order to make my promised numbers. This is completely unethical and in egregio0us cases illegal. But how can you show that management goosed the expected loss? Well, this is just what data did you have and what did you do with it, and an examination of the forking paths you used to get where you ended up…

    • Andrew says:


      Good point. These things come up all the time with the Stan project: we have money in different accounts and with different deadlines, along with lots of uncertainty about future funding. And of course we don’t know what we’re doing, so our accounting’s a mess. It could be a lot worse, though, I’ll say that!

      • Keith O’Rourke says:

        The advice I got for if and when I started a grant funded career was – on the first grant find a way to set aside a part of the funds and only report whatever percentage of the work that is necessary for it to seems like a reasonable first grant success. Use the money and held back work as back-up if the next grant falls short in any way – using FHBFNR accounting (first held back results the first next thing to report). The person giving said that’s how they did managed their career.

        p.s. I started out in accounting in university – the subject is interesting (especially cost accounting – how to assign costs and values) – its the practice year after year that for most does become boring (even at the partner level).

  2. It is interesting how, in many languages, the verb “to count” has the same root as verbs meaning “to tell, relate,” “to reckon,” or “to consider.” Often it’s one and the same verb.

    English: recount/account;
    Spanish: contar (both “to tell” and “to count”), cuento/cuenta;
    Hebrew: spr root (“to count,” “to tell,” and many related words);
    Dutch: rekenen (“to count, reckon”);
    Turkish: saymak, (“to count, take into account, consider, respect, value, deem, suppose”);
    Russian: schitat’ (“to number, reckon, consider”);


  3. R Koenker says:

    A few years ago I had the good luck to hear Elif Batuman give a talk at the British Museum called “Cervantes, Balzac and Double-entry Bookkeeping,”
    which was probably the most entertaining talk I’d heard in a long time, and still resonates with me. The bit on Nabokov’s tennis scoring of Don Quixote
    is priceless, but there are many wonderful connections to many other novels. I think that a recorded version of the talk is still available on web.

    • Yes, I found it, downloaded it, and intend to listen to it on the plane to Istanbul tomorrow! Thank you.

      That reminds me: Luca Pacioli, who wrote the first description of double-entry bookkeeping, also illustrated Da Vinci’s De divina proportione, part of which examines the proportions in the letters of the alphabet. I have a big, beautiful paperback facsimile of the work.

  4. Jonathan says:

    There is speculation that accounting was a big part of the ‘secret knowledge’ attributed to the Jews in medieval into early Renaissance Europe. Don’t know enough about it but there are statutes which made it illegal to use numerals and which required numbers to be written out. The argument then goes that Jews preserved the old methods and the feudal mindset of secrecy and guilds either was extended to that by Jewish merchants or was imposed on them by the Christian world. It makes some sense but then I agree with David Hockney that artists learned how to use lenses and projection and generally kept that secret, even by covering their workrooms in cloaking fabric.

    • jrkrideau says:

      I suspect this “secret knowledge” idea is up there with the story that most people in medieval Europe thought the world was flat before Columbus proved it was not.
      I have only seen one discussion of something like this was in blog posting by The Renaissance Mathematicus, a historian of mathematics and science.

      His discussion of the “the widespread and persistent myth that it is easier to multiply and divide with Hindu-Arabic numerals than with Roman ones would seem to apply to accounting and writing out numerals as words rather than using Roman numerals or later Arabic numerals.

      Boiled down to one sentence. All calculations were done using an abacus so the actual form of how one recorded the numbers was essential irrelevant. There probably was some minor advantage in terms of space and readability but not much.

Leave a Reply