Lesson 3 of 7
Trying to measure the wait
If longer methods yield more, somebody has to say how long a method is. The answer was an average, and it took more damage than anything else here.
By Javier Sanz
Founder of ValueScreener. Built Ninety Nine, a securities services provider, and sold it to a tier-one global bank; ran operations and internationalization at Alpaca, the brokerage API.
Who writes thisEditorial processLinkedIn (opens in a new tab)X (opens in a new tab)
In one minute
- Counting a production process from its earliest input would run back indefinitely, since every tool was made with older tools.
- Bohm-Bawerk's answer is an average: the mean time between when productive effort is applied and when the finished good emerges, weighted by how much effort is applied when.
- The measure is where the tradition is weakest. It was attacked from outside in the 1930s and again in a much larger dispute later, and the attacks landed.
The previous lesson said longer methods yield more. That sentence has a hole in it: longer than what, measured how. Bohm-Bawerk knew, answered it in the same chapter, and the answer has been argued about ever since.
Why the start date cannot be the earliest input
- 1Ore mined, decades agoWith machines made earlier still
- 2Steel madeIn a furnace built before that
- 3Machine tool builtUsing the steel and older tools
- 4Press builtOn that machine tool
- 5This week's outputThe only date that is not ambiguous
The problem, stated
A finished good today rests on effort applied last week, and on a machine built last year using steel smelted a decade ago in a furnace older than that. If the length of a process is the time from its first input, every process is effectively infinite and the measure says nothing.
Bohm-Bawerk's answer
He proposes an average rather than a start date. The measure is the mean interval between the moment productive powers are expended and the moment the finished good appears, so effort applied recently counts for its short wait and effort applied long ago counts for its long one, weighted by how much effort each was. A higher average means a more capitalistic process: more waiting is being carried.
He sets this out in Book II, chapter II of The Positive Theory of Capital, in the same chapter as the roundaboutness claim, because the claim needs it.
Why the measure is in trouble
Three problems, and they compound.
The weights are not observable. Computing the average needs a full accounting of every input and when it was applied, running back through every earlier stage. No firm has that and no statistical agency publishes it.
Inputs are not comparable. Adding an hour of labour to an hour of a machine's service requires a common unit, and the obvious common unit is money, which means prices, which depend on the interest rate the measure was supposed to help explain.
And processes are not lines. Real production is a network in which outputs feed back into inputs, so there is often no single path whose length could be averaged.
What follows for the rest of the track
Take the honest version. The qualitative claim, that production has a time structure and that some activities sit further from consumption than others, does a great deal of work and is not seriously contested. The quantitative claim, that this structure has a length expressible as one number, is contested and did not survive well.
The later lessons in this track are careful to use only the first. When a lesson says a structure of production lengthened, it means resources moved toward stages further from consumption, and it never means a published number went up.
Check yourself
Why can a process not be dated from its earliest input?
- Records do not go back far enough
- Every tool was made with older tools, so the chain has no natural start
- The earliest input is usually land, which is not produced
Show the answer
Answer: B.
The regress is the problem, not the record keeping. It is why Bohm-Bawerk proposes an average instead of a start date.What is the average period of production?
- The time the final stage takes
- The mean interval between effort being applied and the finished good appearing, weighted by effort
- The average age of a firm's machinery
Show the answer
Answer: B.
Book II chapter II defines it as an average waiting time across all the productive powers expended, which is what makes a higher figure mean a more capitalistic process.What is the circularity in computing the measure?
- Output is needed to compute inputs
- Adding different inputs needs a common unit, and using money brings in the interest rate the measure was to help explain
- The weights change each year
Show the answer
Answer: B.
Labour hours and machine services have no natural common unit. Valuing them requires prices, and prices already embed a rate of interest.Sources
- Bohm-Bawerk, The Positive Theory of Capital, 1889, English translation by Smart, Book II, chapter II, where the average period of production is proposed as the measure of how capitalistic a process is.
- Bohm-Bawerk, The Positive Theory of Capital, Books II to V, where the measure is used in the account of how much waiting an economy carries.
What this leaves out
- This is a heterodox tradition. The mainstream account of the same phenomenon differs, and nothing here is a forecast or a view on any security.
- The loaf is an invented illustration of what computing the measure would require. No industry's production schedule is described.
- This lesson states the objections in outline. The originals, including Knight's papers of the 1930s and the later capital debates, are named in lesson seven and are not summarised from their texts here.