Arnold Kling  

What is Economic Activity?

PRINT
More Scott Sumner to Ponder... The Market for Dancers...

Maybe I should write something the length of a Kindle single on Patterns of Sustainable Specialization and Trade. Below might be a way to write the opening.

What is economic activity? In standard macroeconomics, economic activity consists of spending. Certainly that is how we measure economic activity, using national income accounts. However, I propose looking at economic activity as patterns of sustainable specialization and trade (PSST). We can break down economic activity into three components:

Specialization
Exchange
Discovery

Specialization and exchange are the familiar elements of the theory of trade. This is taught as "international trade," using the important concepts of comparative advantage and division of labor. However, trade is trade, regardless of whether international borders are involved.

Economic activity means getting other people to voluntarily do things for you. If I order a pizza delivered to my house, then other people make the pizza and transport it to my door. Last night, when I made a pizza myself, there was less economic activity. However, I still used prepared foods that I bought from a store, including a pre-made crust, a jar of tomato sauce, and already-grated cheese. Somebody else made the oven that I used, as well as the kitchen utensils.

There would have been even less economic activity if I had made the crust myself using flour and yeast, made the sauce from tomatoes and spices, and grated fresh cheese. There would have been even less economic activity if I had used tomatoes and spices that I had grown in the back yard. etc.

I am not saying that doing more of the work myself would have been a bad idea. It might very well have produced a healthier, tastier pizza. But doing work yourself instead of having others do it for you reduces economic activity. From a PSST perspective, this is true by definition. It is also likely to be true from the perspective of the standard paradigm that treats economic activity as spending. That is, the national income accounts will usually show less GDP--less economic activity--when people choose to do things for themselves.

In standard economics, the dynamic properties are thought to reflect investment and capital accumulation. Long-term growth is described in terms of capital accumulation. And short-term fluctuations are often attributed to variations in the level of investment. That was the original idea of Keynes' animal spirits.

With PSST, the dynamic properties reflect discovery. Long-term growth reflects discovery.. (I would not deny that capital accumulation also matters.) Discovery is not so much about "aha" moments or great inventions. I prefer the description in Kevin Kelly's What Technology Wants. In that book, he depicts technology as evolving in a relatively natural fashion. I think that discovery reflects the cumulative application of knowledge.

Part of the process of discovery is figuring out new patterns of trade. Obviously, the transition from the horse to the automobile is going to require many new patterns. So does the transition to widespread use of the Internet.

It is possible for there to be "un-discovery." After the voyages of Zheng He, the Chinese un-discovered ocean-going sailing. Some historians believe that in the Dark Ages, European un-discovered urban living and extensive trade. This may be a stretch, but you might say that financial markets periodically un-discover the rational basis for long-term valuation. I think that governments are quite capable of un-discovering good economic principles.

It is possible that some economic slumps can be traced to un-discovery. However, the more interesting possibility is that slumps can be caused by very rapid technological discovery, with the process of discovering new patterns of specialization and trade lagging behind.

In both standard and Austrian economics, the price system is supposed to take care of the process of adjusting to technological change. This is what we would expect if the instant that a technological change took place, the Walrasian auctioneer quickly tested thousands of different price vectors to find one that induces full employment. In the real world, adjustment is a lot messier.

I wish I could say more. In subsequent chapters I will say more, proposing some crude models. But doing away with the fiction of a Walrasian auctioneer puts me at a disadvantage. It is no longer appropriate to treat the economy as a system of equations with a definite solution. Instead of the Walrasian auctioneer, we have individual business ventures groping for sustainable patterns. Some new ventures will work. Most will fail, as will some existing ventures. There is a tendency for some industries to rise as others fall, but there is no automatic guarantee that the successes and the failures will exactly balance one another. Sometimes, the adjustment can be a difficult one, and the successful new ventures can lag behing the failing old ones.

To put it in terms of a single sentence, slumps can occur because discovery takes time.


Comments and Sharing


CATEGORIES: Macroeconomics



TRACKBACKS (1 to date)
TrackBack URL: http://econlog.econlib.org/mt/mt-tb.cgi/4653
The author at Evolving Economics in a related article titled Kling on patterns of sustainable specialisation and trade writes:
    I have just listened to the recent Econtalk podcast with Arnold Kling on his new “paradigm”, Patterns of Sustainable Specialisation and Trade (PSST). On first thoughts, I am not convinced with the idea. If anything, the paradigm appears to ... [Tracked on February 15, 2011 6:36 AM]
COMMENTS (14 to date)
Sam writes:

"Economic activity means getting other people to voluntarily do things for you."

This is a pretty big shift in definition, no? Traditionally "if a man marries his housekeeper or his cook, the national dividend is diminished"; but under your definition that's still economic activity.

Similarly, if a musician puts up a song for download at no cost and I download and play it, that's not economic activity under the traditional definition, but it is under yours.

fundamentalist writes:

In both standard and Austrian economics, the price system is supposed to take care of the process of adjusting to technological change. This is what we would expect if the instant that a technological change took place…

This is too important to gloss over. Prices do coordinate economic activity very well, but only if they are accurate prices. Accurate prices are those that accurately reflect demand by consumers and scarcity on the part of producers. Socialists conveniently forget that prices can coordinate economic activity only when they are accurate and they are accurate only in a free market. Distorted prices cannot coordinate and the more distorted prices are the less they can coordinate economic activity.

The Federal Register carries an average of 20,000 pages of new federal (not including state) regulations each year. Prices are highly, highly distorted as a result.

The most important price is the price of money. When the Feds pump new credit money into the economy, it distorts the price for which money will exchange for goods and it distorts the interest rate, which is a price for loaned money.

With such massive distortion of all prices by the state, it should amaze everyone that the market works at all.

I’m not a big fan of Schumpeter’s technology theory of business cycles. I still think the monetary theory is the most robust. But if we accept that technology can disrupt price coordination of the market, that again, is a result of price distortions. Here’s why: savings act as a break on new technology. New technology requires increased borrowing to develop and implement. If savings don’t increase, the interest rate will rise and slow down the pace of borrowing for new technology. Thus new technology will be developed and implemented according to the savings rate.

However, if the Feds decide to artificially reduce the interest rate to stimulate the economy, the pace of new technology development and implementation will also increase and cause the distortions in market coordination that Schumpeter’s theory suggests. So it all comes back to money.

Chris writes:

I think you are right to note that discovery takes time, with entrepreneurs trying some successful and many unsuccessful ventures. That is an essential fact about how technological advance interacts with economic activity. However, I'm interested to know what you think about the extent to which, in response to a technological change, the aggregate amount (or length) of discovery is random. Are there so many entrepreneurs responding to the tech change that a sort of law of large numbers applies? In that case the medium-term response to the change is largely pre-determined, though not forecastable. Or can small groups of entrepreneurs have large aggregate impacts because they choose to pursue one set of ventures rather than another set? If that is the case, do you see a role for policy in encouraging more entrepreneurship?

Nick Rowe writes:

The Walrasian tatonnement (groping, trial and error) takes place outside of time, before the market opens for trade. The Klingian(?) tatonnement takes place in real time, as part of the market process. And it takes a long time. And there will be "false" trading, at disequilibrium prices.

Once we put it in those terms, there is less of a difference between PSST and Keynesian/monetary disequilibrium theories of recessions.

fundamentalist writes:

The most egregious example of the failure of prices to clear a market is rent control. If the city sets the rental price above the equilibrium price the market will never clear. That is not price or market failure; that is the government forcing the market to fail for political reasons.

jsalvatier writes:

@ Nick Rowe

Klingian sounds a lot like Klingon.

I hearby dub PSST the Klingon theory of business cycles.

Joe in Morgantown writes:
After the voyages of Zheng He, the Chinese un-discovered ocean-going sailing.

Just a nit: those trips were never PSST.

Philo writes:

"In standard macroeconomics, economic activity consists of spending." This ridiculous canard weakens your presentation.

"I propose looking at economic activity as patterns of sustainable specialization and trade (PSST)." More properly, economic activity *falls into* patterns of specialization and trade, and usually each actor *hopes* that his own pattern is "sustainable" (in many respects; surely not in all).

Everyone knows what exchange and discovery are, but what do you mean by "specialization"? Is this just investing in one's own human capital (perhaps in some rather specific way), or does it include self-advertising--marketing oneself (as some sort of specialist)? And which self-investments count? Surely going to medical school qualifies as "specialization." How about going to college (getting a liberal arts degree)? How about going to high school? Junior high? Elementary school? (Attending elementary school usually transforms one from illiterate to literate. Does this count as "specialization"? Would it so count in a society where only 5% of the people were literate?) The production of a specific good or service is certainly economic activity: You must (rather oddly) be counting this as "specialization," for otherwise it has no place in your scheme.

"Economic activity means [better: is aimed at] getting other people to voluntarily do things for you." But I have heard-tell of "Robinson Crusoe economics"--economics applied to a world containing only one agent. Why exclude this from "economics"? Specialization is something that emerges in a world of many interacting agents, but I do not believe it deserves to figure in the definition of "economics."

"[Y]ou might say that financial markets periodically un-discover the rational basis for long-term valuation." Implausible; EMH is closer to the truth. "I think that governments are quite capable of un-discovering good economic principles." Or maybe they never knew them; or maybe they know them but do not care to act on them.

"[S]lumps can be caused by very rapid technological discovery, with the process of discovering new patterns of specialization and trade lagging behind." Did the pace of technological discovery suddenly pick up in 2006 or 2007? This "explanation" is woefully implausible.

I see PSST as a non-starter.

Dave writes:

"It is no longer appropriate to treat the economy as a system of equations with a definite solution. Instead of the Walrasian auctioneer, we have individual business ventures groping for sustainable patterns."

Gosh, if only there were some way to model activities based individual "agents"... Oh wait, there is, and it's being done on the same campus as you...

R Richard Schweitzer writes:

Your example of spending being taken as a definition rather than a measurement points out where econometrics went off the rails in adapting the work of Colin Clark.

It is as if "If it can't be measured, it should be disregarded."

MernaMoose writes:

fundamentalist,

New technology requires increased borrowing to develop and implement. If savings don’t increase, the interest rate will rise and slow down the pace of borrowing for new technology. Thus new technology will be developed and implemented according to the savings rate.

Interesting. I've read only a little on the money theory.

It seems like the technology development story just isn't simple. For example, the amount of savings "needed" to enable technological progress is not a constant. Some technologies are riskier than others. And some, all risk aside, are just more expensive to implement than others. We aren't always in need of a new trans-continental railroad system, for example.

Savings rate is part of it but, the other really crucial element is how much you're spending on R&D. "High risk" technologies are high risk because, there is still much that is unknown (both on the S&T and the business sides). If little is being spent on R&D then science and technology advance more slowly.

Except that some types of R&D just need a gestation period (my opinion) while new insights dawn on people -- in which case, you can spend too much on R&D. The dawning can take time and I believe, you can't always short circuit the lag by throwing more money at it.

Would you consider R&D part of the "savings rate", or is it something else in your monetary theory? A highly uncertain investment, of sorts?

My thought on the money supply, and savings vs R&D investment rates vs immediate consumption -- there comes a point where you've got enough savings, enough R&D investment, and maybe as a percentage of net economic output it can decline. Because your economy gets so big, that 2% of it now can fund what it used to take, say 8% to fund.

But this is where I always thought the whole "theory" of what's going on, hits an impenetrable fog bank. How do know what's an optimum savings rate? Or the optimum R&D investment rate (let alone the question of what you spend your R&D on)? These are inevitably gambles. For if you save too much and money becomes too cheap to borrow, people will borrow and spend on stupid things (not good). Hypothetically this could happen in the absence of Government Stupidity (yet another inviolate Law of Nature).

R&D -- who knows how you quantify it beyond saying "Wow, it's sure brought us some great stuff in the past". But I've had to make recommendations on how much R&D $ to invest, and when and in what areas, as part of my job. Now if you would kindly point me to the chapter in Economics that tells me how to cut through this fog, I'd be most appreciative. :)

fundamentalist writes:

MernaMoose:

Now if you would kindly point me to the chapter in Economics that tells me how to cut through this fog, I'd be most appreciative. :)

Sorry! Haven't found one yet. I suppose one could calculate an optimum savings rate that would produce the faster economic growth without people starving to death, but it really wouldn't matter because the savings rate depends upon human nature.

Savings depends upon time preference, which is the value we place on present consumption verses future consumption. Time preference changes with age, wealth and other factors. Young people have a hard time saving because they don't care much about the future and want to spend now. Single people and couples without children prefer more present consumption to future consumption. Socialists are very present oriented and care little about saving for future consumption. Social Security, Medicare and Medicaid have all increased our preference for present consumption and reduced savings.

So regardless of what an optimum would be, the nature of the people in society will determine what it is. Then the amount of savings available will determine the rate of technological progress.

Of course, you can speed tech progress up a bit by credit expansion and Hayek suggests that is one of the benefits of credit expansion in his "Monetary Theory and Trade Cycles." But the down side are frequent booms and busts.

As for how much one should spend on R&D at a particular company, the only recommendation I can make is use the old standard from micro econ: invest until the marginal revenue equals the marginal costs. Sorry I can't do better than that.

Colin k writes:

This seems like something to model using Conway's game of life or something else using cellular automata.

It might also be interesting to explore some of the MMORPGs like Eve that have very large and complex economy-like constructs and haven't been too heavily curated.

Steve Roth writes:

I'm quite taken with these sentences, at least:

"What is economic activity? In standard macroeconomics, economic activity consists of spending. Certainly that is how we measure economic activity, using national income accounts."

I repeatedly come across economists who think the measurement is the things itself, even regularly finding that error in economists' actual definitions of the thing.

i.e., since the only way we know to measure utility is what people will pay for it, the definition of utility is what people will pay for something. Things get very circular, very quickly.

BTW, I pulled some data and did come quick calcs a while back, and discovered that if "home work" (i.e. painting your house or repairing your car) were included in the measure of GDP (it isn't, even though it's clearly production), it would constitute about 25% of that total GDP. I think that's an underestimate.

A related question: should leisure time be considered a product? I'm only kind of joking...

Comments for this entry have been closed
Return to top