Why I am not an economist

For graduate students, Paul Krugman is (of course) completely right: if you aren’t very and roughly equally comfortable speaking verbal-narrative, drawing graphical-analytic-geometric, and writing down equation-based mathematical analyses and descriptions of an economic problem, you are doing it wrong and should fix it by learning your tools.

I simply do not think visually. Charts, maybe OK, but graphs? Nope, just don’t “see” them. And equations for me are even worse. I know that some people can read mathematical symbols in the same way that I can and do read a sentence. Ah, yes, if that squiggle moves that way then this other one over here will obviously be doing that. And it simply doesn’t make any sense to me at all.

The only time any academic activity (which may be a sign of how little I’ve done but still) which has literally reduced me to tears is trying to compose equations. People shouting at me that obviously if a is this and x is that then it’s ax2=b or summat. I was able to learn some of the rules about how you manipulate equations: solving quadratics and so on. But how to actually set one up from a description of a problem was and is just entirely gobbledegook.

Err, no, sorry, I just don’t get it.

Which is why I’m not an economist even if I write a lot about the subject, understand bits of it, could even claim to be in a couple of very, very, specific areas out at the leading edge of the discussions of the subject (and grossly and entirely ignorant on many others). My inability at two of these three methods is such that I wouldn’t even actually get into a Russell Group economics course these days. Yes, an undergraduate one. Even back when I was fresh from three or fours years of calculus I wouldn’t get into a course these days. Sure, I knew how to integrate, that can be learned by rote, it’s just something you do with the information provided to end up proving that 1=1. But it never actually made any sense to me.

As equations and graphs just do not today.

In fact, it’s even worse than that. I know very well that I just don’t think visually. Pictures, paintings (no jokes about modern abstracts please), videos, animations, they’re just not the way for me to absorb information.

Odd but there it is.

54 thoughts on “Why I am not an economist”

1. Or if you follow the Austrian School, you understand that arithmetical economics is nonsense, and the equations of Keynes et al are meaningless. And if financial institutions and governments understood that, they would understand why economic disasters keep taking them entirely by surprise. It simply doesn’t work. But drawing lots of nice graphs and writing equations out is very seductive in persuading people that it does. It all looks so darned sciencey, you know.

As the mathematician said to the farmer, “First, consider one perfectly spherical cow…”

2. Only for illustrative purposes and aids to thought. The golden rule is to remember they don’t have any actual numbers on the axes. I think it was Hazlitt put it well; supply and demand curves are mental abstractions which are only of intellectual use as an aid to thought. If you could see them “for real” they’d be snapping around all over the place. They do not actually exist as drawn, and have no predictive power.

3. Surely, isn’t the quality of the pudding in the eating? Economics goes on, whether observed or not. The guidance provided by mathematical economists gives us what we see. Is it, in any way, a success over & above what we could achieve without the guidance?

4. I thought economics was more a mixture of psychology and sociology than maths.

5. Ian B – re spherical cow.
It was a fizzicist. That’s from a very old mathist’s joke.

6. I really struggled with economics at uni, and often for the reasons Tim sets out above. It finally clicked when I saw a graph describing the backward bending supply curve of labour. From that point on, I was on a First.

Having said that, I learn more about economics here, where all the illustrations are done with words.

I suspect that much of the graphs and equations are to forster exclusivity and to clothe the most dismal of sciences in the emperors togs of rigorous academe.

7. Well it’s a bloody good job you aren’t an engineer or a scientist, either. Confessing total ignorance of what underlies modern society puts you (sadly, because mostly I like your stuff) in with the ranks of the Eloi, parasitical on the backs of the Morlocks.

The fundamental problem is that sociology and economics aren’t science or engineering, and for followers of those fields to use the tools of proper science is as daft as (say) a beautiful painting of a suspension bridge supported by balloons …

8. As Ian says, the Austrian school has discovered that reality doesn’t accord with its theories, and has concluded that reality must be at fault.

9. I’m starting to wonder if Ian is actually Ritchie, because the population of people able to so spectacularly misunderstand the subjects they pontificate on must be pretty small.

As for graphs and equations, a graph is just a visual representation of an equation, and an equation is just a more concise way of writing a sentence that could be written in words. If you’re having trouble changing the words into the equation, it’s because you haven’t actually got the sentence right; it’s much easier to miss something, or gloss over it, in a sentence than in an equation.

That we’re having this conversation at all is a damning indictment of the way maths is taught in schools.

10. @Rachel Scotland

“Confessing total ignorance of what underlies modern society puts you (sadly, because mostly I like your stuff) in with the ranks of the Eloi”

I read the post as saying that he’s not a visual thinker, not that he was utterly ignorant of economics.

And doesn’t your post assume that the tools of ‘proper’ science could describe economics perfectly, like if only those thick economists got out the way and let intelligent people have a go?
Nice thing about physics and engineering is that there are demonstrably true answers. Not the case with things that describe people’s behaviours, rather than universal laws.

11. The list of things that “the tools of science” can neither describe nor control/manipulate is a very long one.

Does Rachel Scotland choose her relationships –of whatever form–by “scientific” means?

SJW–And when– not if– your gang discover people don’t match how socialists think they should be the left murders them. Don’t think you have a lot of room for calling out Austrian economics.

Still waiting for that “inevitable” working-class revolution aren’t you. Except you know its not coming now so the white working class is to be destroyed by immigration instead.

Dave–your grasp on reality is not sufficient to be talking about what others don’t understand.

12. John Square,

“Nice thing about physics and engineering is that there are demonstrably true answers. Not the case with things that describe people’s behaviours, rather than universal laws.”

I think economics is an immature science and very hard to measure due to complexity and confusion. But are people looking at actions and outcomes, forming hypotheses and getting broadly the right answers? I mean, I’m not even an economist and I can get right the outcome of raising the minimum wage. I can’t say exactly how many jobs will be destroyed, but doctors can’t predict the performance of a cancer drug on cancer cells for a particular body, but we call that science.

I sincerely believe that the alternatives to neoliberalism will be seen like creationism in a few decades.

13. Before the quantitative comes the qualitative. The numbers simply support, but do not explain, the idea.

Not everything can be or is best explained by numbers.

The effects of a prism can be explained without the complex maths needed to prove quantum theory of light.

If you can explain concepts of economic in terms that non-economists can understand without all the charts and numbers, it is a gift.

Stick at it.

14. Not everything that counts can be counted; not everything that can be counted, counts.
A. Einstein (attrib)

15. Abstruse macro-Economic “science” maybe, probably is, useful for central banks and so forth, but Economics will not tell you how to run an economy, or how to improve it. If it did, all governments would be the same.

There are two flaws: 1) the dependence on “everything else being equal”, which it never is, almost by definition, and, 2) entrepreneurial types tend to be done with school by about 14, so never learn this stuff anyway.

A minimum wage will destroy some jobs, everything else being equal. What the overall effect is, who knows.

16. No, Dave.

A sentence is something like “increasing the money supply will tend to cause inflation”.

An arithmetical economist tries to write “increasing the money supply by delta x will increase the aggregate price level by delta y” but that is the part that does not work.

If it did, the economy would be predictable and controllable. It is neither of those things.

17. No Ian

Mathematical economists are well aware that the economic world is complex and uncertain. Do you know what the S in DSGE stands for?

18. Has anybody put mathematical economic models in front of Murph for him to describe and discuss? Or even better, teach, in his capacity as 0.2 Professor?

19. A lot of economics envy here today. I can’ t do it so it must be shite.

If you whingers are so much better, where’s your prizes for showing it? Your outstanding much cited papers that show that your theory is better.

And it’s no good saying it’s “immature”. We still have to try. Or do you think remaining ignorant is a better solution?

You have my sympathy Tim. Despite the slur on modern education above (unwarranted, since you weren’t educated recently), some people just can’t see things visually or in equations. It’s good that you just accept it, rather than asserting all equations must be worthless. I love that sort of thing. My weakness is aural — can’t do music, accents, languages etc. It is what it is.

20. SJW-

It ought to stand for “shite”. Elaborate bunkum is still bunkum.

21. At school I was given Carl Jung’s autobiography to read. The bit I remember is about how he didn’t get algebra. I can’t say it made my teenage self think well of him…

22. I find it a bit difficult to get people who don’t get algebra. I took to it very easily. But I think it’s one of those basic brain wiring things. It’s like there is a certain degree of abstraction at which each brain gives out and falls into bamboozled mode, and there’s not much you can do to improve upon that. It might be down literally to the number of neural connections available to particular brain cells required for that bit of thinking, and once they’re all in play, that’s it. No more abstraction possible.

23. @ SJW and Ian B
To a large extent, maths is something you are born with or without. I was born lucky but I (have) know(n) some perfectly decent intelligent guys who just couldn’t do maths. Was I supposed to look down from my ivory tower – what? – no!- they were just as good as me as people: they just couldn’t do maths.

24. @ Chester Draws
I did actually get a prize (for which I had not knowingly entered) from one of my professional bodies for something I had written which had economic implications. However the poor quality of my singing and foreign accents has resulted in hundreds or thousands of disparaging comments.

25. Anyone who writes any books with graphs or the like should be forced to read Edward Tufte first.

I think economics has suffered serious Physics Envy and gone down the path of mathematics. Which is a mistake. Their maths is poor and poorly understood. And it probably doesn’t actually work in most cases to describe what they think it describes. It provides a feeling of confidence that is misplaced. Sure, if they want to provide the current value of a bond, fine. If they want to model the real world, not so much.

26. Someone some day ought to compile a list of books recommended by TW’s blog readership. I assume it would start with Oliver Rackham’s books on trees.

But Edward Tufte really ought to be there.

27. If the maths used in macro-economics is wrong, it’ll be because the information needed is just not there. Replacing math with sentences isn’t going to make a blind bit of difference because the issue if not the usefulness or otherwise of maths per se, the issue is surely the attempt to predict and to some extent control the aggregate outcomes of human behaviour – for which the information is insufficient.

You can’t make a silk purse out of a pig’s ear.

28. The problem is mostly definition of terms.

To illustrate; consider the assertion that women find confident men more attractive. This appears to be a reasonable assertion. But the mathematical economist then writes “attractiveness is proportional to confidence” and A=kC.

But you cannot assign a numeric value to either variable, let alone derive a constant of proportionality. You simply cannot do arithmetic with this, even though you can write these things down.

Keynes does the same type of thing over and over again, for instance. His terms (such as investment) sound clear enough, but there is no way to properly assign numbers to them. Sometimes he means savings, sometimes he means government spending on anything, sometimes he means government spending on “infrastructure”, sometimes he means any spending on infrastructure, sometimes he means any spending on “investment goods” and so on and so on.

29. Ludwig Von Mises said:

As there are in the field of social affairs no constant relations between magnitudes, no measurement is possible and economics can never become quantitative.

That’s pretty powerful if true. Think of all the labour that could be saved.

30. I find it a bit difficult to get people who don’t get algebra.

I know a Russian guy who learns languages in a matter of months. His latest was Farsi. He’s fluent in Hindi. He studies dead languages for fun. He probably doesn’t get people who don’t get languages. As you say, it’s a brain wiring thing.

31. Nice thing about physics and engineering is that there are demonstrably true answers.

Sort of. But a lot of “engineering” these days is getting something done in a certain set of conditions. Sure, the calculations need to be correct and demonstrably true, but past that an awful lot of it becomes subjective (and usually it comes down to cost).

32. If you can explain concepts of economic in terms that non-economists can understand without all the charts and numbers, it is a gift.

I quite agree. Going back to engineering again, one of the biggest problems the really good engineers have is how to stop being an engineer for a second and do something else. Most of the good engineers – meaning, those who can do the calculations and engineer the technical solutions – can’t do stuff like organise, present, record, report, set a priority, manage interfaces, etc. I figured out very early on in my career that I was a rubbish technical engineer but pretty good at all the other stuff that needs doing but engineers are often not much good at. Although I get a bit embarrassed as sometimes I feel like a complete fraud: my contribution being to organise a few meetings, get a budget together, and kick a few obstacles out of the way of the discipline engineers and not much else.

33. But Ian that’s two problems. One is the assignment of numerical values, and the other is the definition of terms to which numerical values are being assigned.

34. “Sure, the calculations need to be correct and demonstrably true, but past that an awful lot of it becomes subjective (and usually it comes down to cost).”

I’ve often wondered about that when considering the choice of spillway design at reservoirs, e.g. whether or not to build tainter gates or open lip overflow designs. If you have a gate, you can alter the opening aperture to let more or less water out. Less water lost means more money saved. If you build an open lip overflow you have no such control. But it’s probably cheaper to build and there is the added benefit of knowing the dam’s continuing existence doesn’t depend on adequate gate maintenance and engineers not falling asleep or being otherwise incapacitated.

35. As someone with no visual memory(I can have an hour long conversation with you and if we meet somewhere unexpected, even 5 minutes later I will not recognize you) math tends to come easily to me. I attribute this to the fact that in order to remember what is going on in the world I have to use words or numbers. Sometimes words are more descriptive but at other times a bit of simple math works best. I do tend to be more qualitative than quantitative but when you are suck using only words or numbers then space for details can quickly run out.

Earlier minimum wages were mentioned. When I look at the pretty graphs of GDP and overlay the timing of minimum wage increases I get the feeling that an increase in minimum wage generally is followed by an increase in GDP. The problem in getting the quantitative analysis correct is making sure to apply all of the appropriate variables. At the most basic level wage increases should lead to a loss of jobs for the lowest income workers. I haven’t seen this to the extent that would be expected if this is the only effect so there must be more. One factor to consider is that in the US only Democrats tend to vote for minimum wage increases. Severe economic crashes tend to occur more frequently when Republicans are in charge so it is possible that the GDP increase is simply a normal business cycle. Another option is that increasing minimum wage means that people with lower incomes have added disposable income which gets spent quickly. The McDonald’s worker now can afford a few more luxury items at Walmart which helps to offset the added wage costs. Obviously these are only 3 of possibly hundreds of variables we actually need to consider. The problem with math in economics isn’t the math but the people who write incomplete equations. I have no way to quantify what the actual effects of the wage increase are as neither I, or anyone else, really know exactly what the 187th order effect is that needs to be in the formula.

I guess according to Krugman I can never be an economist since I can never master the visual aspect of his pretty pictures. The fact that his math is incomplete must not matter in his mind. I think I’ll ignore that tripe and do my best to understand a subject that is more convoluted than quantum mechanics.

36. I’ve often wondered about that when considering the choice of spillway design at reservoirs, e.g. whether or not to build tainter gates or open lip overflow designs. If you have a gate, you can alter the opening aperture to let more or less water out. Less water lost means more money saved. If you build an open lip overflow you have no such control.

You look at two things, really: CAPEX (cost to build) and OPEX (cost to operate/maintain). Once you’ve done a few similar projects you start to standardise designs because the experience tells you that one works better than the other in practice.

Unless you’re a French company, in which case all experience can be ignored in favour of the whim of the incredibly clever manager at the head of the table who believes he has found something which the previous eleven generations of engineers have overlooked. This is labelled “optimisation”, and the result is every single design is bespoke and key features are selected by individuals who won’t be on the project in a year’s time, and no record made of the decisions.

37. Although I get a bit embarrassed as sometimes I feel like a complete fraud: my contribution being to organise a few meetings, get a budget together, and kick a few obstacles out of the way of the discipline engineers and not much else.

But I bet all the other engineers are bloody glad you’re doing all that so they don’t have to.

38. But I bet all the other engineers are bloody glad you’re doing all that so they don’t have to.

Exactly.

39. You should get a job in the NASA manned space programme, then you could spend all your time designing things that will never get built anyway, with an occasional press release naming a random future date for going to Mars.

40. “You look at two things, really: CAPEX (cost to build) and OPEX (cost to operate/maintain). Once you’ve done a few similar projects you start to standardise designs because the experience tells you that one works better than the other in practice.”

Well that kind of makes sense; in the reservoirs I’m looking at the older ones (built in the 60s and 70s) tended to use tainter gates and the newer ones (80s, 90s and 2000s) tend to use open-lip overflow. What’s baffling are the exceptions, e.g. one built in the south as recently as 1995 with three relatively small tainter gates – after the trend toward open-lip overflow had already been established. The really early ones (20s and 30s) were all overflow or hydraulic gate designs. There have been no dam failures meaning all the spillways are functioning like they are supposed to, so it’s the switch to tainter-gates in the 60s and 70s that has to be explained. Are engineers prone to fads?

41. Are engineers prone to fads?

Very much so. Bear in mind that engineers are human like everyone else, and the same bullshit office politics and power games go on around the calculations and drawings. So some dickhead comes in and wants to show his management that he is “adding value” and so dicks with a proven design, and then points to this (preferably highly visible) design change as “his mark”.

Plus, it’s amazing how much can come down to the simple preference of the guy designing something: I asked a structural engineer why he had designed a whopping great structure with 4 legs when every other one in existence had 3, and he looked at me blankly and said “Well, why not?”

42. Haha, because they were dickheads… That does at least have the virtue of being humorous as well as explanatory.

43. @ Liberal Yank
” When I look at the pretty graphs of GDP and overlay the timing of minimum wage increases I get the feeling that an increase in minimum wage generally is followed by an increase in GDP. ”
That is due to one of the failings of the way GDP is measured – goods are included at purchase price, but government functions are included at cost. So an increase in public sector pay appears to increase GDP.

44. @john77
What does public sector pay have to do with minimum wage?

At least in the US public sector wages do not get an automatic bump just because minimum wage goes up. The only wages affected directly are those at the very bottom of the economy, generally restaurants and retail.

45. @ Liberal Yank
Who cleans government buildings? Who employs care workers to look after pensioners on social security? Which sector is unionised so that wages go up every time minimum wage increases push up the cost of living index?
Are you tell me that the private sector can hire cleaners for less than the government in the USA?
Blair’s New Labour “outsourced” cleaning to nominally private sector companies that took over the labour force and a contract to do exactly the same job to make it look as if the public sector workforce was not growing, but as soon as the minimum wage went up so did the cost of cleaning government offices and so did GDP without so much as a single extra doughnut being baked and eaten.

46. SMFS, who hasn’t got a clue, tells us that economists’ maths is poor. Ian B, who doesn’t recognize the word ‘stochastic’, asserts that all economic models are bunkum, principally because they assume all relationships are deterministic.

In the face of such powerful arguments, I find myself firmly on the side of the economists.

47. @john77

Cleaning is done either by government employees that start out at well over minimum wage or by a contractor. The contract, unless it happens to run out just as a minimum wage increase goes into place, won’t change immediately. In neither case is a minimum wage increase a first order effect. Wages, in the US government at least, have steady COLAs that are applied yearly. Once again a minimum wage increase will have 0(zero) first order effects. Yes the private sector does hire cleaners for less than the government. They are called illegal immigrants who get no benefit from minimum wage increases as they are paid, illegally, less than that amount.

There is so much that is wrong with your second paragraph that it barely merits a response. Are you telling me that none of the workers who now make a little more bought a single extra doughnut? That claim seems insane to me. Please show me evidence that a worker that makes more doesn’t spend at least part of that wage increase.

We can debate whether concentrating wealth in the top 1% or distributing it more evenly results in better growth. Your argument appears to be that a minimum wage increase only drives up the cost of government without any of that money being recirculated. If the money earned by those that get a raise directly due to minimum wage increases does not get respent where does it go? Are poor people literally burning it just to watch it burn? Even if it is just burned doesn’t that count as entertainment?

48. Social Justice Warrior – “SMFS, who hasn’t got a clue, tells us that economists’ maths is poor. Ian B, who doesn’t recognize the word ‘stochastic’, asserts that all economic models are bunkum, principally because they assume all relationships are deterministic. In the face of such powerful arguments, I find myself firmly on the side of the economists.”

Being habitually beaten by people he regards as intellectually and socially beneath him, Paul resorts to an ad hom to explain away why he is habitually beaten by people he regards as intellectually and socially beneath him.

Myself, I think that there are some economists who would welcome such a high powered intellect on their side. They would probably approve of his intellectual integrity too. Maybe SJW can become a 0.2 of a Professor if he works hard?

49. @ Liberal Yank
Cleaning staff get paid at the end of the week. So they didn’t buy any extra doughnuts until they got the cash in their pockets – after the rise in GDP I mentioned.
And the increase in production of doughnuts would have a time-lag, only following an increase in full-price sales to the point where the number of doughnuts sold off cheap at the end of the day was negligible.
Get real!
In your textbook there is an infinitely small increase in production every rtime there is an infinitely small increase in demand, but in real life production increases in steps – the only economical method of baking doughnuts is in large batches and selling off cheap at the end of the day those left on the shelf.
You are falling into the Murphy fallacy of the Magic Money Tree (MMT) by assuming that the added income of the cleaning staff is not matched by a fall in income of someone else. If, and only if, the propensity of cleaning staff to spend their extra income is not matched by an equal drop in the spending of those whose income is decreased do we get the secondary effect that you claim.
It is certainly *not* insane to claim that reported GDP goes up before the workers spend a penny of their pay rise because that is an elementary piece of algebra.
There is a case for claiming that there will be a secondary, and real, increase in GDP as a result of a rise in NMW because the propensity to save is, on average, higher for those with higher incomes. The argument almost always ignores those who become unemployed and have a substantial fall in income as a result – regular readers of Tim’s blog are bored with my citing the evidence that a rise in NMW causes a major rise in unemployment among low-wage workers in the traded goods sector – and it seems these unfortunates have escaped your scrutiny.