David Hume’s nightmare

Here’s David Hume from “Of the Balance of Trade

The skill and ingenuity of EUROPE in general surpasses perhaps that of CHINA, with regard to manual arts and manufactures; yet are we never able to trade thither without great disadvantage. And were it not for the continual recruits,° which we receive from AMERICA, money would soon sink in EUROPE, and rise in CHINA, till it came nearly to a level in both places. Nor can any reasonable man doubt, but that industrious nation, were they as near us as POLAND or BARBARY, would drain us of the overplus of our specie, and draw to themselves a larger share of the WEST INDIAN treasures.

Ha!  Soon after we stupidly invent container ships the Chinese buy up half of central London.

PS.  Over at Free Exchange, S.C. made this comment:

But durability is not quite all there is to it. A car, for example, provides a useful service. It also lasts more than a year. If a firm adds a car to its fleet of vehicles, it counts as investment. But if a household buys a car, it counts as consumption. Why? The short answer is the obvious one: it was bought by a household not a firm. . . .

So Mr Sumner is not quite right. Durability is not the only thing that counts. Distance from the perimeter of production can also come into it. Investment, we might say, is the construction of durable assets. Comma. The purchase of some durable goods (cars, washing machines, and so on) can count as consumption, not investment, if those goods are sufficiently close to the consumer–if the services they provide are enjoyed directly by households, rather than being embodied in another good, mediated by a market transaction, or otherwise well captured in the GDP statistics. There is a bit of “give” in how statisticians think about these things.

I agree.  But surely what we are interested in is not consumption and investment as measured by government statisticians, but rather the concepts that are of interest to economists concerned with evaluating public policies.  Admittedly the distinction between consumer and capital goods is a matter of degree, but the criteria that matters is durability, not how the government measures or mismeasures it. Perhaps I’ve misunderstood the entire point of Goldman Sach’s exercise.  If the only goal was to show that Chinese investment may be more helpful to consumers than many critics have assumed, then I’m on board.


Tags:

 
 
 

32 Responses to “David Hume’s nightmare”

  1. Gravatar of Saturos Saturos
    8. October 2012 at 10:02

    So Hume didn’t understand comparative advantage? I guess no one did before Ricardo…

  2. Gravatar of Saturos Saturos
    8. October 2012 at 10:03

    Scott, the Economist has a whole topic heading for “Scott Sumner” now. And Tyler Cowen has cited you as an authority on Korean film on the blog, so watch out…

  3. Gravatar of Greg Ransom Greg Ransom
    8. October 2012 at 10:14

    Hume was wrong about learning, causation, perception, language .. and this.

  4. Gravatar of ssumner ssumner
    8. October 2012 at 10:50

    Saturos, I recall he argues that at some point the loss of specie would lower European prices enough so that trade would again be balanced. It’s actually a monetary essay, not an essay on comparative advantage. Can anyone confirm?

    Greg, No one’s perfect, but Hume is damn close.

  5. Gravatar of Saturos Saturos
    8. October 2012 at 10:53

    That’s what I thought the essay was about too, but here it sounds like he was saying that Europe could never profitably trade with China…

    The Goldman Sachs excercise is just The Fatal Conceit. Current or capital doesn’t matter, without a free market there’s no way to tell whether the most valuable outputs are being produced.

    I agree with Scott about Hume.

  6. Gravatar of ssumner ssumner
    8. October 2012 at 10:53

    Saturos, What do you mean by “topic heading for Scott Sumner?”

  7. Gravatar of Saturos Saturos
    8. October 2012 at 10:55

    Scroll down to “Related Items” under the Free Exchange post you just linked to.

  8. Gravatar of Saturos Saturos
    8. October 2012 at 10:56

    Or just look here: http://www.economist.com/topics/scott-sumner

  9. Gravatar of ssumner ssumner
    8. October 2012 at 10:56

    Saturos, Maybe the “overplus” of specie referred to the extra money Europe had got from the Americas that made its per capita money stock higher. It wouldn’t lose all money, just that extra amount.

  10. Gravatar of Saturos Saturos
    8. October 2012 at 10:59

    Well, I’d like to think Hume independently discovered comparative advantage… but then why wouldn’t he tell Smith?

  11. Gravatar of Doug M Doug M
    8. October 2012 at 15:37

    What is Hume talking about….

    China had all sorts of stuff that Europeans wanted to buy. However, the Chinese showed no interest in purchasing any of the manufactured goods that the Europeans were trying to sell. For example, clocks were vital for the budding industrial revolution. They were the the high tech of the 18th century. The Europeans were emensly proud of their clockworks. European ships would come to China to try to sell clocks and the chinese could not have cared less.

  12. Gravatar of Doug M Doug M
    8. October 2012 at 15:44

    What is investment? Durability is neither necessary nor sufficient. Inventories are investment, and not necessarily durable. Jewelry is durable, but it is not “investment.”

  13. Gravatar of Patrick R. Sullivan Patrick R. Sullivan
    8. October 2012 at 16:14

    Along came Jones;

    http://www.econtalk.org/archives/2012/10/garett_jones_on_2.html

    Scott gets his name mentioned several times.

  14. Gravatar of Philemon Philemon
    8. October 2012 at 18:58

    +1 @Doug M

    Up till the mid 18th century, Late Imperial China enjoyed a tremendous balance of trade in its favour in silver against Europe. She produced a lot of stuff highly sought after elsewhere, but weren’t sufficiently interested in what the Europeans produced–beyond the silver (mined in Peru). Fancy clocks were the rage among the rich, but there isn’t enough demand for it to make enough of a difference. Strict restrictions on who when where the Europeans were allowed to trade also did not help and might well be a bigger factor than any inherent in ability for the Europeans to profit from comparative advantage. But once the British discovered that the Chinese would buy opium and subsequently ramped up production in India, the flow of silver soon reversed. The rest, as we say, is history.

  15. Gravatar of Major_Freedom Major_Freedom
    8. October 2012 at 20:37

    ssumner:

    Saturos, I recall he argues that at some point the loss of specie would lower European prices enough so that trade would again be balanced. It’s actually a monetary essay, not an essay on comparative advantage. Can anyone confirm?

    It’s an essay within a collection of essays on political economy. The essay ‘On the Balance of Trade’ contains Hume’s “famous” hypothetical example of a sudden fivefold increase in the supply of money and how this would, in a world market, lead to an outflow of money to other countries due to competitive balancing forces.

    Admittedly the distinction between consumer and capital goods is a matter of degree, but the criteria that matters is durability, not how the government measures or mismeasures it.

    Durability is not the criteria that matters, for many reasons. For one thing, it is an ill-defined concept. There is no physical law of the material universe, nor is there a logical category of human action, that would convey to us that good X is durable whereas good Y is not durable. If you say a durable good is one that lasts more than X years, then this would require us to call a (X minus one day) good a non-durable good. Regardless of what time frame we choose, there is no rigorous foundation for why that time frame rather than another time frame, is justified.

    The reason why so much confusion arises in analyzing the concepts of consumer goods and capital goods, in distinguishing them, and in clearly setting boundaries that serve as criteria for nomenclature, is due to the fact that most economists, and most sources of economic theory, do not ground economic theory on human action, but rather on some empirical foundation, something “out there”, in nature, and the belief is that just like physicists learn what things are by considering a thing’s nature, so too can economists learn what things are by considering a thing’s nature. But this is a mistake. Economics doesn’t study things. It studies us humans, specifically, our actions. Thus, consumer goods and capital goods have to be defined and understood by being grounded in humans in some respect, not “the thing in itself”.

    A good’s durability cannot be the foundation for identifying a good to be a consumer good or capital good, because there are both short and long lived consumer goods, and both short and long lived capital goods. Durability will only lead us to considering durable consumer goods and durable capital goods. It won’t lead us to separating consumer goods from capital goods.

    There are many capital goods that most would consider to be legitimately defined as “short lived”, or “non-durable”. For example, milk that is bought by a bakery, for use in the production of cakes, is short lived. But milk for a bakery is a capital good, because the bakers don’t consume the milk to satisfy a direct end, i.e. their direct utility, as a consumer good would, but rather, the bakers productively use the milk in order to produce a good for sale to others. This, as I am sure you will agree, is a good that serves not the baker’s direct end, not their direct utility, not their consumption desires, but an indirect end, an indirect utility. Milk to a baker is a capital good.

    It doesn’t matter that the milk is used up very quickly relative to other capital goods (like the ovens), because there is always a more durable good we can compare a given good to. Relative to the oven, a factory is even more durable. So here, the oven would be “short lived” relative to the factory.

    Time, in economics, is a praxeological concept. It is not objective. “Durability” only makes sense within praxeological constraints. A good is durable because it is more long lasting than other goods. But there is no strict line that says shorter than X is a consumer good, and longer than this is a capital good.

  16. Gravatar of Saturos Saturos
    8. October 2012 at 21:43

    The state of public economic understanding: http://www.helium.com/debates/672476-do-rising-home-prices-mean-the-economy-is-improving

  17. Gravatar of Lorenzo from Oz Lorenzo from Oz
    8. October 2012 at 22:18

    Would people stop with the “Chinese weren’t interested in what Europeans had to sell” crap. Folks, it’s elementary economics. Outside sub-Saharan Africa, the medium of exchange was silver. The Chinese economy was possibly still larger than the European economy; at worst it was comparable in size. Silver went from the Americas to Europe to China (some went directly from the Americas to Europe). In all, about a third of American silver ended up in Asia/China, so two-thirds stayed in the Atlantic economy.

    So goods in the Americas were very silver dear, goods in Europe were comparatively silver dear, goods in China (and the rest of Asia) were silver cheap. Of course Chinese (and other Asian) goods went to Europe and silver went to China (and rest of Asia).

    The minute the Europeans came up with a good where demand was relatively inelastic (opium), they were away. Then the Industrial Revolution happened and suddenly European goods were much cheaper and people stopped talking about silver drains and Chinese “not being interested” in Western goods.

    Conversely, silver was not the medium of exchange in (West) Africa. So the Europeans could sell goods to Africa, buy slaves, ship them to where land was cheap and labour was expensive (the Americas) and ship silver-expensive American goods (often produced by said slaves) to Europe.

    I fail to see what is at all mysterious in any of this. Or why we have to appeal to mysterious Chinese dislike of Western goods; they were too bloody expensive given that silver was the medium of exchange, that was the problem.

  18. Gravatar of mbk mbk
    9. October 2012 at 01:58

    “The Chinese”, if we so wish to aggregate across 1.3 billion people, seem to be quite interested in Louis Vuitton bags and Mercedes cars. Just sayin’.

    I am still scratching my head as to the odd notion that “investment” should somehow be linked to “durable”. This has popped up before on this blog, I still don’t understand the discussion. Maybe it’s a macro thing.

    In business and investment circles an “investment” is an expense that is expected to yield a return. “Consumption” is the purpose and final goal of previous “investments”: an expense not expected to yield a monetary return.

    Thus:

    A car bought for going to and fro a job is an investment.
    A car bought for weekend pleasure rides is consumption.

    Milk bought for baking and selling the baked goods is an investment.
    Milk bought for drinking oneself is consumption.

    etc.

  19. Gravatar of Saturos Saturos
    9. October 2012 at 04:23

    IMF thinks “the fiscal multiplier” is >1, not 0.5: http://www.ft.com/intl/cms/s/0/16771cb2-113b-11e2-a637-00144feabdc0.html#axzz28ht2sfpD

    mbk, definition of investment is different between economics and finance. See any good intermediate macro textbook, eg. Mankiw.

  20. Gravatar of Saturos Saturos
    9. October 2012 at 04:52

    What do we think of Yichuan Wang’s new slides on NGDP targeting?
    http://synthenomics.blogspot.com.au/2012/10/nominal-gdp-targeting-introduction-with.html

    He’s presenting at his university’s finance club on Thurdsay.

  21. Gravatar of J.V. Dubois J.V. Dubois
    9. October 2012 at 05:17

    MBK: I know, it is very confusing. I was there. There are actually three different languages: Macro, Accounting and “Average Joe”

    In Macro, from what I understand the Investment is defined as “not-consumption” in a little complicated way. Consumption is defined as purchase of newly purchased goods and services consumed during a defined time period (usually a year). The rest is marked as “saving”. And saving equals investment (it is an identity). Just as an interesting note, given sufficiently small time period (like the second after you recieved your salary), all your income is saved. Given sufficiently long time (like millions of years), all your income is consumed as decay is the nature of the world we live in.

    So in between these two extremes we have a fuzzy space where people get confused. So there can be many views of what is considered as investment and what is not. Most importatnly, any durable good can be considered as a capital good that produces services in the future. Everything from cars, home appliances, chess sets, clothes etc. produce services: transportation, refrigeration, chess games or “protection from element”. People derive utility mostly from these services, not from holding the goods that produce these services.

    So for instance a chess set is consumer durable that can survive for centuries providing countless chess games to many generations of its owners. But from accounting point of view it is hard to calculate the value of this stream of chess game services provided. That is why accountants just use shortcut and pretend that the chess set was consumed the day it was sold to household and that it will provide exactly its market price in terms of chess games.

    But this is still important even if it is not measured by official GDP statistics. But there are important innovations that fall into cracks of national accounting and that are very important for the standard of living. Just one example – ebay. Being able to sell old stuff that you no longer need (you derive no utility from services this good provides) to somebody who values it increases social welfare. While it does not necessary increase the capacity of the society to produce new goods every year, it gets more bang for everything that is produced. Reusing and repairing old stuff, or producing (and consuming) goods that are build to last frees resources that can then be used in other areas. This in other words mean “saving” which is “investment”.

  22. Gravatar of mbk mbk
    9. October 2012 at 05:24

    Saturos,

    mbk, definition of investment is different between economics and finance. See any good intermediate macro textbook, eg. Mankiw.

    Having Mankiw at hand let me quote pp497-498 of “Principles of Economics”, N. Gregory Mankiw, 6th International Edition: “When you hear the word investments you might think of financial investments […] By contrast […] here the word investment means purchases of goods (such as capital equipment, structures, and inventories) used to produce other goods”

    It turns out my definition and examples cover both the macro textbook definition (“goods purchased to produce other goods”, such as my car and milk) and the business use of the term (“an expense that expects a monetary return”). These two definitions are quite similar in nature anyway and neither invalidates my critique of the spurious term “durable” in this context.

  23. Gravatar of mbk mbk
    9. October 2012 at 05:32

    JV Dubois,

    I didn’t even consider the I=S identity which really is a special case that’s only used in a certain kind of macro model. Either way it doesn’t say anything about durability as you explain.

    But here is my point: redefining terms that have a modicum of precise meaning in a practical sense (accounting, investment, macro a la Mankiw), in ways that the new definition can mean practically anything depending on context (here: time scale or “durability”), is one of the hallmarks of cults. I just can’t take this seriously in any practical way.

  24. Gravatar of J.V. Dubois J.V. Dubois
    9. October 2012 at 05:59

    MBK: Saving must equal investment, if saving is defined as any income that is left after purchasing consumption, and income is defined as … Ok, lets not go this way.

    I think the problem is that economy uses words that already can have different meanings in common language. Or that people do not understand that the meaning of some words changes if viewd from 10k feet – like the word “Income” that in Macro simultaneously stands for “Expenditure”.

    Anyways, to distill what is important is this: imagine that people do not consume goods, but they consume services. At the very moment that some of your need is satisfied, you just consumed some service provided to you by someone or something. In this setup, capital good is anything that stays there after providing the service and that has capacity to continue providing this service in the future.

    The fact that you cannot easily assign market monetary value to the flow of these services (and thus to the capital good itself as the source of these services) it does not make it less capital.

    PS: I do not understand your cult reference. We live in a fuzzy logic world where many definitions are fleeting. Take a very known example of “bald man paradox”. If you start with head full of hair and then you lose one hair does it make you bald? If you lose another hair will you become bald? If there is not exact distinction between “hairy” and “bald” does it mean that I cannot take seriously anyone who uses either of these words because they are not precisely defined and they don’t have universally accepted meaning?

  25. Gravatar of mbk mbk
    9. October 2012 at 06:57

    JV Dubois: I could have worded this better, but I don’t quite agree that economics uses words differently from everyday language as a rule. Business, accounting, finance, and Prof. Mankiw’s textbook on Economics all give pretty similar definitions for “investment”. Mankiw’s definition as I quoted above in particular is quite unlike the Keynesian I=S identity, and it is drawn from a passage where he explicitly talks about measuring GDP (hence macro). Mankiw doesn’t mention durability at all.

    Now for sure I=S has its place in a particular well known model but the passage by “S.C.” quoted in Scott’s post, mixes up this particular macro model with everyday definitions AND with the idea of durability, and as a result, IMHO, it becomes a complete mess.

    So: I’d argue that the I=S model should not be used in conjunction with any “real life” train of thought, involving real stuffs bought and sold by real people and companies. And I’d argue that bringing in durability in the way you argued above, and in the way Scott argued before in this blog, and again above, has no explanatory value for economic decision making. It makes a previously clear distinction used by professionals in the economy (expected ROI: works both for perishable milk and durable cars) into an unclear one used only by the economics profession (variable durability as a yardstick). Pun intended.

    But even if you don’t believe me: how to explain Mankiw then? And why can’t we have macro models that make sense to accountants, businessmen, households and bankers, all of which use Mankiw’s definition?

  26. Gravatar of J.V. Dubois J.V. Dubois
    9. October 2012 at 07:41

    MBK: Maybe it is me who does not get it. Mankiw’s definition is all about Keynes – he defines the word “Investment” in a way it is used in national accounting. There investment is defined as sum of business fixed investment, residential investment and inventory investment. This definition is useful from the point of view of national accounting.

    But even with this definition you can expand it beyond this narrow national accounting meaning. If investment is spending on things that are used to produce other goods or services, then purchasing new oven so that you can bake your own bread should also count as investment. Don’t you think? If you follow this train of thought, you will find out that it is the ability of some good to be used to generate stream of future services that really makes it as valuable investment. Not unlike residential housing that is used as the source of residential services for households is regarded as investment.

    If there are two countries with similar marketable production capacity and in the country A people purchase more cars and home appliances and other durables that will last for years then the next year they will be potentially “richer” then people in country B where everybody had expensive vacation last year.

  27. Gravatar of Doug M Doug M
    9. October 2012 at 07:44

    I=S only works if you take a vary broad view of S that would include corporate retained earnings, government surplus and foreign flows as savings.

  28. Gravatar of Saturos Saturos
    9. October 2012 at 08:14

    mbk, in the sequel “Macroeconomics” he distinguishes between “investing” in IBM shares and actually constructing new plant and equipment. Economic investment increases the capital stock.

  29. Gravatar of Saturos Saturos
    9. October 2012 at 08:16

    J.V. Dubois, yes textbooks tend to fudge the different senses of the term. Hence my definitions: http://www.themoneyillusion.com/?p=16425#comment-186777

    (Actually there’s a mistake there, it should read:

    Investment: The allocation of production to productive capital goods, including the accumulation of unsold inventories of output which will deliver future benefits.

    )

  30. Gravatar of mbk mbk
    9. October 2012 at 09:15

    J.V. Dubois, I know I am tilting at windmills because at least in this blog Scott’s usage of the term is as you describe. I know this. Iam just trying to point out that this is not the way pretty much everyone in the economy sees it, save professional macroeconomists. Example, we are here debating some Goldman Sachs paper. Or some “investment” report. All these people talk about “investment” in a clearly commonsensical, Mankiw-esque way. Now if one takes this common sense, corporate-accounting, or IRA-accounting context, and then starts using the word “investment” indiscriminately for items such as say, wristwatches (because they’re durable and provide time-keeping services), one is bound to run into confusion with people. This is what I am trying to point out.

    By the same token, if one redefines “investment” as “anything that provides a future service” then yes, everything can potentially be labelled an “investment”, because everything provides some kind of service. If everything is an investment then the term has lost its meaning. Or rather, the meaning becomes arbitrary. Should I invest my money in an IRA or in a Porsche? No difference: both are clearly investments. No wonder pundits are confused, the public is confused, professional investment advisors are confused. Everyone is confused. One of the major terms in the equation can mean anything one sees fit.

    Saturos, whether it’s “investing” in IBM stock or “investing” in new plants for factory production: both examples still clearly fit the idea of “expense with the expectation of future monetary return” so it doesn’t matter which one you use – it’s still not the “I=S” kind of definition and it still is completely independent of durability. In both cases the money still goes towards producing further goods yet to be sold to end consumers, albeit in a slightly different way.

  31. Gravatar of J.V. Dubois J.V. Dubois
    9. October 2012 at 10:11

    MBK: I understand what you are saying. Most economist use the word “Investment” in national accounting sense: that investment is any spending by firms on items not incorporated within sales.

    But this is just very narrow meaning of the word. For instance, it would imply that prior to late medieval age – when incorporation was invented – there was no investment. It was just “households” like farmers, craftsmen and nobles purchasing “consumer durables” like plows, anvils and castles.

  32. Gravatar of Major_Freedom Major_Freedom
    9. October 2012 at 10:34

    mbk:

    I am still scratching my head as to the odd notion that “investment” should somehow be linked to “durable”. This has popped up before on this blog, I still don’t understand the discussion. Maybe it’s a macro thing.

    In business and investment circles an “investment” is an expense that is expected to yield a return. “Consumption” is the purpose and final goal of previous “investments”: an expense not expected to yield a monetary return.

    Thus:

    A car bought for going to and fro a job is an investment.
    A car bought for weekend pleasure rides is consumption.

    Milk bought for baking and selling the baked goods is an investment.
    Milk bought for drinking oneself is consumption.

    etc.

    I use the same criteria as you do for distinguishing capital goods from consumer goods. I find it to be the most consistent, and most pragmatic. One, it does not rest on the physicality of the thing itself, which is a necessary criteria, since economics is not chemistry, but the study of acting man; two, it relates to money expenditures, which is crucial for understanding the economics of a division of labor society.

Leave a Reply