The Curious Case of Missing Global Productivity Growth
economist.com
economist.com
For example, you could look at solar PV where last year the industry set a record of 70 gigawatts-peak of modules installed. The price competition is so fierce that manufacturer revenue has barely budged over 2012, when only ~30 GWp were installed, and actually declined from 2015. If an economist looks at this sector do they see spectacular productivity improvements like I see (far more watts, at lower prices, with real physical improvements up and down the supply chain), or do they consider productivity "missing" because the physical improvements aren't leading to revenue/profit growth? And if it's the latter, does that imply that economists only recognize productivity improvements when markets are imperfect enough that improved production technique delivers more money instead of fiercer competition? Is there no such thing as growing productivity (that would be recognized by mainstream economists) in a deflationary environment?
But that doesn't get clicks.
https://en.wikipedia.org/wiki/Output_(economics)
Makes sense so far. A rich country is one that has abundant goods and services for its people, regardless of how that abundance comes to be. But then I read the examples below the introduction and everything's measured in money. Which do economists consider a more fundamental measure of Output -- increasing access to tomatoes, or increasing the revenue of the commercial tomato sector? If the latter, that's nuts. It implies that all sorts of smart investments in efficiency and/or self-sufficiency are unproductive or even destructive if they lead to less future paid exchange of goods and services.
(1) Inflation correction. If technology makes tomatoes cheaper, now dollars are worth more, counted in tomatoes. Technology is almost by definition deflationary. Economists do try to correct for deflation and inflation in these statistics, though everyone admits it's not possible to do a perfect job. I think this correction partly addresses your concern about measuring goods vs dollars. Inflation measurement tries to keep the ratio of tomatoes to dollars accounted for, so that when we measure dollars, we can do inflation correction to translate into tomatoes.
(2) Elasticity. If the cost of tomatoes goes down, then presumably the quantity of tomatoes produced will rise in response. This may increase the overall value of tomatoes produced (though it also may not). Over the arc of human history, technology that makes things cheaper to produce has tended to stimulate overall production. E.g., we work longer hours than many hunter-gatherers.
(3) Objectivity. Many economic measurements fail to get what after we really want, which is how much happiness are we producing. But since happiness measurement is even more fraught with error, counting up widgets is hopefully not too bad a substitute. Although it is theoretically flawed, in practice I think it clearly succeeds. The people in countries with high GDPs seem more prosperous than the people in countries with low GDPs.
But yeah, GDP is flawed. A state with awful weather where everyone needs to work to air condition their homes will have a both a higher GDP and a lower quality of life compared to a state with great weather.
This is an attack on he article, but the article made quite clear that measuring productivity is notoriously challenging! You didn't have to look far - just the first sentence of the second paragraph:
> Measuring productivity is far from easy; it tends to be the residual left over when all other factors have been accounted for. The OECD says it "can often be a measure of our ignorance".
You've accused the authors of the article of something here (your vague insult means you can pretend you meant whatever you want after the fact - no point in my guessing what you were accusing). But what I'm seeing is that you responded to a comment without reading the article, and still chose to insult the author of the article, and the publication itself to boot.
You could have just answered the parent poster's question directly rather than randomly lambasting media.
But that doesn't get upvotes.
Empire State Building according to Wikipedia: $645 million in 2017 dollars
One World Trade Center: $3.9 billion
And that is before we look at things like worker safety, bureaucratic demands etc.
We do harder things now.
The Bank of America Tower, for example, has about the same floor space as the Empire State Building, and "is considered a worldwide model for green architecture in skyscrapers", yet cost $1B.
Even if you want to throw in the $120 million for fixing the Empire State Building's energy efficiency in 2009, it still doesn't say great things for modern productivity.
I think your question is asking whether we are mismeasuring the value of dollars over time. And certainly there are worthy criticisms of inflation measurement, and good reasons to think that they overstate inflation as a result of undermeasuring quality improvements. But these flaws are not unique to recent history, so by themselves, they cannot explain why productivity growth has ostensibly slowed.
One way to sidestep the issue of statistical dollar measurement is to just read direct history of past time periods. Doing so has given me a much richer appreciation for the radical transformations caused by industrialization, urbanization, electrification, etc. A terrific article on this subject was published last year in the New York Times, comparing the years 1870, 1920, and 1970:
https://www.nytimes.com/2016/05/15/upshot/what-was-the-great...
The article was inspired by Robert Gordon's lengthier book, The Rise and Fall of American Growth, which followed a somewhat similar structure:
His descriptions of 1870 really made me appreciate how truly impoverished the average person was in those days. The rise in quality of life as electricity and automobiles and running water percolated through society over the next 50 years was radical, far more radical (in my view) than getting computers and the internet and phones, as monumental as those achievements have been.
Ultimately, it's impossible to objectively compare value across different situations. But my own semi-informed take is that productivity growth is indeed slowing.
My impression is that many commoditized products continue to see decent year-over-year improvements in the ratio of inputs to outputs. But if economists don't care about material input:output ratios, only currency-values, then improvements-plus-fierce-competition could look like stagnant productivity. You would need process improvements plus GDP growth to recognize productivity gains; flat GDP per capita and simultaneous shrinking inputs of matter and energy wouldn't count.
However, this could apply equally to previous time-periods. Just as modern generations benefit from superior electronics products and mobile network speeds, previous generations benefited from massive increases in the variety and freshness of food, or the household appliances that reduced the time cost of household chores. It could be that productivity growing at a rate that is both faster than measured but slower than trend?
What you mean to say is 'can we measure consumer surpluses' instead of 'producer surpluses' - i.e. profits.
Indirectly - yes - this may be the root of the problem.
When products improve in quantity and quality, but profits go down - we generally say the 'economy is shrinking'.
But the value to consumer has gone up.
We don't measure 'consumer surplus' i.e. the value captured by a consumer when they buy something.
Essentially - you pay $1 for a pen. How much would that pen cost so that you felt 'it was not worth it'. Well, you might pay $100 for a pen, if you need to write sometime in your life. You'd probably treat it very well, but you'd still need to buy it.
Maybe $500 is the real value of a pen. In this case - there is $499 'consumer surplus' every time a pen is bought for $1. (Again, not quite fair as you might buy several $1 pens whereas you'd only buy one for $500 - but you get the picture).
The GDP is just the cost of everything sold in the economy + government spending.
Which is a really bad measure if you think about it :)
- also measuring inputs in these terms (e.g. wages paid to labour, rents paid to capital)
- adjusting prices with reference to a selected base-year so that movements in MFP/TFP are comparable across years
And empirically, what has been historically observed is actually the opposite of what you suggest: productivity and inflation are inversely correlated (i.e. periods of high inflation have show low productivity and vice-versa). Here's a paper from some absurd and hubristic mainstream economists on this issue: https://www.rba.gov.au/publications/rdp/2003/pdf/rdp2003-10....
It's actually even more complicated than just this. But I can't really be bothered to get in to it. But good luck with that Mises Institute stuff.
From a theoretical standpoint, productivity is a volumetric measure (Outputs / Inputs). From a more practical standpoint, even if output is measured in dollar denominated terms, the methodology used to measure total/multi factor productivity (TFP/MFP) minimise the effects of inflation on the calculated residual, primarily by:
- also measuring inputs in these terms (e.g. wages paid to labour, rents paid to capital)
- adjusting prices with reference to a selected base-year so that movements in MFP/TFP are comparable across years
And empirically, it has been historically observed that productivity and inflation are actaully inversely correlated (i.e. periods of high inflation have show low productivity and vice-versa). This is what you'd generally expect if MFP/TFP estimates are largely unaffected (directly) by inflation.
The inverse correlation is likely due to the efficiency sapping effects of high inflatation (which imposes frictional costs and mispricing inefficiencies) at the 'whole of economy' level. And at the industry level, it's likely a mix of the latter and, in cases of exogenous output price increases (e.g. booming commidity prices), due to firms throwing more inputs at producing more marginal outputs (which is now a profitable thing to do due to higher sale prices). Here's a paper on this issue: https://www.rba.gov.au/publications/rdp/2003/pdf/rdp2003-10....
A similar, somewhat controversial example of a "distraction when math tries to becomes words" can be seen in the news when it comes to discrimination. Studies of sociodemographic differences and income levels, after controlling for education and IQ and experience etc., find differences in pay between men and women and between different minorities. The unexplained remainders for differences in income got called discrimination. Discrimination is a real thing and a real problem and must be addressed. But the reasons for differences in pay is not entirely due to discrimination, it's clearly much more complicated than that, studies that show this complexity are only recently getting attention.
Lumping and labeling a problem as one big thing ("technology", "discrimination") that's easy to explain to a layperson but when it's really much more complex and the problem still largely misunderstood is a disservice to everyone.
TFP does not do that. It doesn't measure anything. TFP is a residual from theoretical economics, not empirics, often styled in a form of Y=AKL. We don't know how (or at least can't agree how) to measure K or L accurately, more importantly we don't even know what A is or what it contains.
I am not surprised you find TFP useful. It's a Nobel Prize winning idea. And I respect the thinking behind it. But it has since been kindly debunked by the real world. Many people find prayer or homeopathic medicine useful and practice them daily, this doesn't mean those things are working how people think they are or at all.
I don't mean to punt an answer here but Wikipedia explains the problems with TFP better than I. Follow the links through the Cambridge Critiques as a start: https://en.wikipedia.org/wiki/Total_factor_productivity
For measuring productivity gains from technology, what does "go study its real impact in the real world" entail? Is this the current thinking of state of the art productivity/tech economics?
How do you disaggregate one technology from another in a system as complex as a business, let alone an economy? These are the hard (impossible) questions that TFP addresses - we don't try to look at each part individually in a vacuum, we impute the total effects of "technology" and try to control for 2-3 main variables: capital, labor, and output elasticities.
But today, the availability of empirical real world data or a way to go get it if not yet collected is night and day different compared to 60 years ago when many of the old economic theories were being thought up and became popular. Go read the old papers there is no data just theory. Today, economic theories and formulas can often very quickly be tested with empirical research and data. And in my opinion current economics has turned toward empirical work having weight over theoretical work
How do you disaggregate? You do the best you can. And yes we do try to look at each part individually (not in a vacuum though). Because it turns out aggregating is worse. When you look at TFP the way you describe, you are not just aggregating technologies you are aggregating everything...aggregating all technologies and also whatever else is happening that K and L didn't capture. Who needs that, it doesn't tell you anything practical but that "other stuff is happening". Technology might turn out to be only <5% of TFP. There's enough evidence that TFP has too many "factors" in it to be lumped as just technology. Physicists didn't give up when they realized 95% of the universe is dark matter/dark energy that they don't understand, economics is following that lead, double down on the small stuff and hope to understand the big stuff a bit better and the best you can.
I looked at your bio, not sure if you use TFP in your work or its just a personal interest. I know a bit about about market and equity research. If you have access to the data I would use PCA over TFP, if you don't have this I would just proxy a series and risk off systematic variables and focus on key known numbers. Better to ignore fake data than listen to hurtful real data. Either way we are all always guessing.
https://en.wikipedia.org/wiki/Productivity_improving_technol...
https://en.wikipedia.org/wiki/Aggregation_problem
Blah blah blah: https://www.youtube.com/watch?v=VVp8UGjECt4
But I do find TFP to be a very interesting concept that captures a whole ton of important externalities and fundamental infrastructure/inputs to production at the macro level. I was just wondering if you could point me towards something that would do a better job of that than TFP, because I haven't come across one yet in my studies.
An alternative theory posed by one of my favorite economists[0] is that administration of public companies under the "maximize shareholder value" doctrine is finally coming home to roost. By prioritizing short-term profits in favor of long-term investment, companies have slowed their own growth for the last ~30 years.
[0] http://www.goodreads.com/quotes/434485-running-the-company-f...
Of course, that does not mean that the robots won't take all the jobs on the near future. Just that they didn't do it already.
I didn't dig deep enough on it to make my mind, and there are plenty of reasons to think it's flawed.
However the more likely answer is that the measurement is wrong.
Or perhaps there is less hope or belief that working hard will result in getting ahead. With most of the gains post-crisis going to the top and income inequality growing, perhaps there is less worker incentive to work hard.
2. In cities such as NYC (where I live), Boston, DC, SF, LA, ...there is rent-seeking in the real-estate market through zoning density restrictions causing market inefficiencies that increase the cost of housing far higher than in a functioning, efficient market.
Harvard Economist Edward Glaeser has written about this in the context of affordable housing in NYC, in the article, Build Big Bill [DeBlasio, NYC Mayor] [1]
3. There are other cases of rent-seeking throughout the economy as well as other market inefficiencies through negative externalities and information asymmetry. Reasons why healthcare in the US is so expensive is negative externalities of smokers, air pollution, not paying their way for additional healthcare costs and information asymmetry where doctors have much more information than patients and patients may have much ore information than insurers related to medical conditions.
[1]http://www.nydailynews.com/opinion/build-big-bill-article-1....
What if Productivity was driven by demand?
Do we "produce" stuff (or offer services) without any expectation to sell it to someone? If something is selling like hotcakes the producers of such item will do all they can to increase productivity, and competitors will try to create a similar product which either costs less or offers more features (or both) in order to capture the unfulfilled demand.
Ok, assuming this is reasonable, what happens when demand, in general terms, starts to lag (due to a long economic stagnation/crisis)?
(Another symptom that could help strenghten my argument: Hanjin built a large shipping fleet expecting to need it to ship stuff around and cope to increasing demand... and now: http://www.abc.net.au/news/2016-09-05/south-korean-shipping-... )
Otherwise, how would you redistribute the assets? Gold bars can be shipped, but land ownership is difficult. For example, would a random illiterate from a developing nation be assigned ownership of a Manhattan skyscraper? One could imagine a system of guardianship in place to steward the property, but that system would enable a new aggregation of wealth through fees.
It seems to me that your proposal would be an extremely efficient way for a small group of people to rapidly consolidate wealth.
Sounds like capitalism.
e.g. before Docker/Containers came along, tools like Chef/Puppet/Ansible were already automating much of infrastructure deployment/management. But a new paradigm was introduced which was (seemingly) better and that is catching on now.
Take Wikipedia, for example. Back in the day, a set of encyclopedias cost hundreds of dollars. Now Wikipedia is free (if you don't donate to them), and has more information. That produces value to a huge number of people. But how much does it show up in GDP? I suspect it shows up as zero dollars.
Or take Linux. Or Google. Both of those contribute far more value than they earn in dollars.
I suspect that there's a lot of that in the information sector - more value than GDP being produced. And this makes us "richer", not in dollar terms, but in terms of information and tools available for us to use.
I spent a few years self-employed, I would bid a money cost for an accomplishment based on how many hours it would require. If my bid was accepted, I was then immediately motivated to find a more efficient approach so I had more of both time and money.
Every iteration of that approach improved my ability to accomplish tasks in smaller amounts of time.
https://www.lordabbett.com/content/dam/lordabbett/en/images/...
I wonder whether goods becoming more durable in general is an influence. Cars last a lot longer now than when I was young, but they aren't as easy to make. It's easier to churn out shoddy goods than long-lived ones; as consumers insist on better quality, the "same" item takes longer to make.
First: total_factor_productivity = price_of_output/(0.7labor_cost + 0.3capital_cost)
Let's say we're in a state of ideal competition - the minute some company improves productivity , which leads to reducing the input costs they put into the process by X% , other companies copy that process, which lead to the price going down by X%.
And on the other hand, companies insulated from competition usually don't grow their productivity.
So maybe the economy has many more of those 2 types of companies and that's mirrored in the economic statistics ?