We should ditch GDP as a measure of economic activity
This article exposes the false economic concepts behind GDP, which is only the visible tip of a large iceberg of economic deceit. Describing an increase in GDP as economic growth owes its meagre validity to imprecise definition. An economy does not grow, only the quantity of fiat currency deployed grows. A successful economy progresses our condition, our wealth, our standards of living. The evolution of misleading statistics such as GDP to their current condition is only governed by their usefulness to governments, not as an objective development of sound theory by seekers of truth.
There are perhaps two plausible reasons for producing the GDP statistic, other than employing statisticians, and both have nothing to do with economics. By compiling the figures, a government keeps track of its tax base, and it can enter into the game of my-country-is-bigger-than-yours.
In international comparisons of economic performance, gross domestic product adjusted for price inflation is the most common metric used. Countries are ranked by size, and success is measured by the rate of growth in GDP. This is important to the political class.
About two years ago, I was told that the Indonesian central bank had a plan to do away with cash entirely, because it would bring unrecorded transactions into Indonesia’s GDP, promoting it from sixteenth to perhaps the thirteenth largest nation measured by GDP. I have no idea if this was true, but allegedly, this was important to the Indonesian government.
We should not be surprised if going cashless is partly motivated to give the illusion of GDP growth, in the same way that in 2014 the EU decided to add in estimated contributions from prostitution and drug dealing. These are examples of why and how GDP is manipulated to produce a goal-sought answer.
As a widely-accepted measurement of economic activity GDP is a fairly recent introduction, dating from the 1990s. Before GDP, we used gross national product (GNP) which was introduced following the Second World War. And before that we had national income, a theoretical concept that goes back to Sir William Petty in the seventeenth century.[i]
The more you examine the history of attempts to define national income, the more you see that it has been adapted to suit the mathematical economists of the last century in their statist objectives. While attempts are made to include or estimate black market activities, second hand goods are excluded, which make up a significant part of our economic activity. In its construction and purpose GDP has amounted to compromise heaped upon compromise. That notwithstanding, it is now the principal method of comparing the economy of one country with another.
The relative status in a table of size measured by GDP is always rationalised into US dollars as the common reserve currency, along with the implied rate of comparative growth. This cuts across domestic prices, which corrected for purchasing power parity gives completely different answers. The rate of growth from one year to the next is adjusted by the rate of price inflation in the underlying currency to come up with “true” GDP growth. One wonders at the purpose of it all, given, as Petty points out in his Political Arithmetick, the size of a country is immaterial to its wealth, which is better judged on a per capita basis. The Indonesians in the example above would be better off considering how to promote individual wealth than to manipulate statistics to enhance their international standing.
This article looks at the principal deficiencies of the GDP statistic, and why it is more misleading than useful. Erroneous beliefs that GDP records economic reality could make the task of central banking’s crisis management more destructive than it would otherwise be, a looming issue given that the global credit cycle is moving closer to an inflection point triggered by rising interest rates.
This article questions the legitimacy of inflation-adjusting, the diversion of attention from the importance of payments in the production process and makes the case for excluding the government sector entirely. It also shows why it is wrong to make any association between an increase in GDP and economic progress.
The senselessness of inflation adjustment
In an inflationary economy, official government price statistics routinely under-record the true rate of price increases approximated by raw data. By this simple device a GDP increase is registered when the economy might be going nowhere or even regressing.
The exaggeration of real GDP growth is greatest in a high-inflation economy, but it is also evident when there is greater price stability. US Nominal GDP in Q2 2018 grew by 7.6% annualised, while the CPI is recorded as rising at an annual rate of 3.2%. Therefore, the official growth rate of the US economy for that quarter was 4.2%.
Independent sources agree the rate of price inflation in the US is closer to 10%.[ii], in which case adjusted for true consumer price inflation, GDP contracted by approximately 5.8% annualised instead of increasing. This may be difficult to comprehend, because we are so conditioned to official statistics that always say otherwise. Yet we are prepared to accept black market prices as the true indicator of the rate of inflation in high-inflation economies, such as Venezuela or Argentina, because official statistics paint too rosy a picture. But it cannot be one rule for the reserve currency and another for some of the others, because that is simply inconsistent.
Assuming the independent estimators of US price inflation are correct, it also throws the whole basis of monetary policy into doubt. If in Q2 2018 real GDP contracted by 5.8%, it implies to policy-makers that interest rates are far too high to stimulate the economy and should be cut sharply. Yet, if the Fed is targeting price inflation, interest rates are far too low and should be increased to 10% just to be neutral.
The answer to this conflict is straight-forward. The real world works on nominal prices, not inflation-adjusted ones. A businessman constructing a business plan will consider known factors in calculating his returns. He starts with an assumption of the volumes he can expect to sell based on known market prices. He then estimates the costs involved. From these costs and the market value of his anticipated production, he can work out the expected return on his capital. Very rarely do these estimates incorporate some form of price indexation, unless prices are inherently unstable. And even then, rather than indexing future prices between the business planning process and eventual production output, he is more likely to increase the hurdle rate for returns on capital.
It is nominal prices that matter. Economists are mistaken in their attempts to adjust GDP by price inflation, not least because today’s price inflation is the result of monetary expansion long ago. But there is also a fundamental misconception of what GDP actually represents, which is easy to demonstrate.
Over the course of a year, the total income earned on wages, rents, interest and profits spent in the domestic economy is, by definition, GDP. In an economy where the total amount of money and credit is unchanged from one year to the next, total income must also be unchanged, and therefore all else being equal, so is GDP. Meanwhile, the economy progresses. New technologies and production methods are invented, and consumer demands evolve. All forms of capital, including labour, are constantly redeployed to improved effect. Individual prices rise and fall, driven by changes in consumer demand. All this happens despite there being no change in the total national income, and therefore there is no change in GDP.
The central point is that GDP does not in any way measure changes in the economic condition.
Now let us consider an increase in the quantity of money deployed in that part of the economy deemed to be included in the GDP statistic. Hey presto! We have economic growth. If five percent of the sum of base money and bank credit is so applied, GDP rises five per cent. Therefore, increases in nominal GDP are simply the effect of monetary expansion.
Logically, if there is to be a deflation of GDP, it should not be by a price index, but by the quantity of the extra money involved. That would reduce “real” GDP growth back to zero, and in effect express GDP in the prior year’s money. And we can all agree that is a pointless exercise.
Therefore, it is an egregious error to try to deflate GDP by a rate of price inflation. Everyone does it, and central banks run monetary policy in the belief that it is appropriate. It is small wonder central banks are economically destructive with this level of logical comprehension.
The error in passing GO
There are two main approaches to the calculation of GDP, expenditure-based or income-based estimates. The former is of final expenditures (the money spent on goods and services) and the latter of the total income earned on wages, rents, interest and profits.
In theory, both approaches should amount to the same thing. At this point, we must dismiss a common assumption behind GDP, that it is only final expenditure or income that matters. The error has a long history, and goes back even to Adam Smith, who wrote,
“The value of goods circulated between the different dealers never can exceed the value of those circulated between dealers and consumers; whatever is bought by the dealer being ultimately destined to be sold to the consumers.”[iii]
In the sense of the sum of added values in the production process, this is obviously correct. But more relevant in a true estimate of economic activity is payments, payments in the production chain and payment for the final product. The payments between the various stages of production are usually a multiple of the final value, a fact which is hidden from us by the GDP statistic. Indeed, it is only partly revealed to us by the business-to-business activities that occur in production as captured by the US Bureau of Economic Analysis’s gross output statistic (GO).[iv]
This simple fact shows that as a measure of economic output, GDP only tells part of the story and misses a substantial chunk of economic activity. In the US, gross output in manufacturing is estimated by the BEA to be an additional 1.75 times final values. Manufacturing, which tends to have a higher element of gross output than services, is about one half of US private sector GDP. In a manufacturing-based economy such as China Germany Japan or South Korea, it will be even higher.
Gross output was only recently introduced by the BEA after lobbying by economist Mark Skousen, who was merely following the Hayekian tradition.[v] Hayek’s triangle illustrated the role of payments at different stages of production, and how they would alter when the balance between savings and consumption changed. Hayek showed that the relationship between final products and payments in their production and delivery are far from static, and if accurately known could be a useful leading indicator, far more effective than GDP.[vi]
The economic establishment still fails to get it, seemingly unaware of Hayek’s triangle because it is ignored in mainstream literature. Particularly risible is the footnote to the GO statistic on FRED, the St Louis Fed’s statistical data base, which describes GO as “According to the source [the Bureau of Economic Analysis], gross output is….”
In other words, FRED does not appear confident enough to define GO itself. There’s little doubt that the confusion is one of mainstream economists being wedded to Adam Smith’s definition of the value of goods, and not appreciating that the economics of production involves considerably more than final values.
On this, statistics have displaced reason. The more you look at it, GDP is as far removed from representing economic activity as Earth is from Mars.
Government activities should not be included
Back in the late nineties, when Gordon Brown was Chancellor of the Exchequer in the UK, he could always demonstrate in his budget statements that the economy had grown more than generally expected. The way he achieved this was simple: he spent more money on government services, and it was the growth of this spending which inflated the growth rate of GDP.
Today, government spending represents an even larger share of nearly all national economies. It is funded by taxes, debt and monetary expansion, all extracted from the private sector. Taxes remove wealth and income from the productive population. Debt subscribed to by ordinary citizens directs their savings to generally unproductive use. Debt subscribed to by banks is inflationary funding by the expansion of bank credit. What the government does with the money extracted from the productive economy has three important consequences: it destroys wealth, debases money, and it produces little or nothing.
Given the point of GDP is to measure economic output, it is clear that to include government spending, and the income of those employed by government and its agencies, is not in the spirit of GDP. Furthermore, as noted above government debt is funded wholly by the private sector, and the inclusion of government in GDP numbers gives a false impression of its true debt position. Adjusting the GDP numbers by taking government out of them dramatically changes a government’s financial situation, as Table 1 below illustrates.
For the welfare states, the relationship between government debt and its funding source is revealed as truly grim, while every Western economists’ bête noir, the Chinese government, suddenly looks relatively good. However, not even these figures address the private sector’s ability or willingness to cover outstanding government debt, or a government’s ability to collect the higher taxes to pay it off.
If the economic community was honest, it would recast the relationship between GDP and government spending to exclude it. By including it, the myth of government being a productive force is perpetuated, and by excluding it, the cost of government to national production is highlighted.
The GDP statistic is an abuse of economic theory. Very few people seem to understand that it is not a measure of economic growth (or more properly expressed, economic progress), but simply a money-total. Attempts to turn it into an inflationary-adjusted statistic by applying a deflator are simply wrong. It is just a money total.
It doesn’t even include the majority of productive activity, which is better reflected in the actual payments, or gross output, involved in bringing a final product to market.
And lastly, government spending should be excluded on the grounds that it is always at the expense of sequestering the income of genuine producers and destroying the accumulated wealth created by the private sector, producing little or nothing in return.
[i] See Petty’s Political Arithmetick published in 1676. Petty listed the trades and commodities exchanged to compare the wealth (incomes) of Holland, England and France.
[ii] See Shadowstats calculation based on 1980 methodology: http://www.shadowstats.com/alternate_data/inflation-charts and also Chapwood Index at http://www.chapwoodindex.com/
[iii] Wealth of Nations, Book II, Chapter 1.
[iv] See the BEA’s explanation of GO: https://www.bea.gov/faq/index.cfm?faq_id=1034
[v] Skousen lectures on economics and finance at Chapman University.
[vi] See Roger Garrison’s article at https://www.auburn.edu/~garriro/b3beyond.htm
Find more: Contributing Authors