Calculating GDP CorrectlyTags Other Schools of Thought
While gross domestic product is considered by many to be the most important statistic regarding our economic well-being, Austrian and non-Austrian economists alike have criticized it as being an unsound representation of the health of an economy. Readers of mises.org are probably familiar with these arguments and I am not going to rehash these critiques.
However, some of the criticisms of this statistic misrepresent how the statistic is determined. I think it would be instructive to explain how GDP is calculated.
GDP is the federal government’s monetary estimate of all final goods and services produced in the country in a specified time, usually annually or quarterly. They calculate GDP using the expenditure method—in other words, they add up all the categories of spending on final goods and services to calculate GDP.
Here is the formula for GDP:
GDP = C + I + G + NX, where
C is personal consumption,
I is gross private domestic investment,
G is government purchases,
NX is net exports. Net exports equal exports minus imports.
The GDP total of consumption spending, investment spending, government purchases, and net exports for 2019 was $21,429.0 billion. The table below provides details of this data.
Personal consumption is divided into purchases of consumer goods and purchases of consumer services. According to the 2020 Economic Report of the President, in 2019, we purchased $4,508.6 billion of goods and $10,055.2 billion of services for a total personal consumption of $14,563.9 billion. This was roughly two-thirds of the total GDP for 2019.
The second category of expenditures is gross private domestic investment. Gross private domestic investment is gross in the sense that it includes all investment, private in that it does not include any government spending, and it is domestic investment meaning that it includes all investment spending in the US.
There are two ways for a business to invest in itself: it can increase its capital stock and it can build up its inventories. Gross private domestic investment comprises fixed investment, investment in capital stock, and changes in inventories. For 2019, fixed investment totaled $3,742.8 billion. $2,878.7 billion of this total was business investment, called non-residential investment. Interestingly, 35 percent ($1,012 billion) of the non-residential investment was investment in intellectual property. According to federal statistics, intellectual property is a major portion of the capital stock of the US.1
They also consider spending on new housing to be an investment. The remaining portion of fixed investment was $797.4 billion in residential investment.
Now we get to a rather trivial portion of the GDP statistic, but one that has led to some unfounded criticisms of the GDP calculation, changes in inventories. For 2019, we had a $66.8 billion change in inventories. That means that total business inventories at the end of the year exceeded those at the beginning of the year by $66.8 billion. It makes perfect sense to add the increases in inventories to GDP. Consider a new truck built in the US in 2019. The truck was produced but not sold in 2019, so it was added to some company’s inventory. Since we want to include this truck as a part of 2019 production, it is appropriate to add increases in inventories to GDP.
However, some have concluded that subtracting a decrease in inventories from GDP leads to a flawed statistic. That is not the case.
If inventories decrease during the year, we subtract that decrease from GDP. For instance, in 2009, changes in inventories equaled –$150.8 billion, showing that inventories decreased significantly that year.
At first glance, it does not make sense to subtract anything from production. It is impossible to produce a negative amount of anything. You cannot produce negative ten trucks. Nevertheless, consider a truck that was produced in 2018 but sold in 2019. The truck was included in the 2018 GDP and was in inventory at the beginning of 2019. When the 2018 truck was sold, the 2019 inventories decreased. The appropriate method would be to omit this expenditure as a part of 2019 GDP since the truck was included in the 2018 GDP. That is exactly what happens. When the 2018 truck was sold, it was added into 2019 GDP as either a consumer good, a business investment, a government purchase, or an export. Since they do not want to count this truck in the 2019 statistic, they subtract decreases in inventories from GDP. All decreases in inventories are added to some other category of GDP and then subtracted from GDP in the changes in inventories category. In this manner, they ignore all previous production when calculating the 2019 GDP.
There is one more issue regarding inventories that needs to be addressed, how intermediate goods are treated when they are added to inventories. Intermediate goods, the goods that are inputs for final products, are not generally included in the GDP statistic. GDP measures the production of the final goods and services produced in an economy and it would be inappropriate to include intermediate goods in this calculation.
However, intermediate goods that are added to inventories are included in GDP. If some of the parts of a truck, the engine and transmission, for example, are produced in 2019 but not assembled into the final product, then those intermediate goods are added to the company’s inventory and are included in GDP. When these 2019 intermediate goods are used to produce a truck in 2020, the intermediate goods are subtracted from inventories at that time.
Government purchases are straightforward. Government purchases are comprised of federal purchases, $1,423.4 billion, and state and local purchases, $2,330.8 billion, for a total of $3,754.3 billion. The federal purchases are divided into two categories, national defense, $846.6 billion, and non-defense spending, $576.8 billion.
There are two interesting items in these numbers. Federal spending for 2019 was $4,448.3 billion (this does not include the $11 trillion spent paying off the bonds that came due), but only 32 percent of this spending was on the actual purchase of any goods or services. Moreover, although defense spending was a relatively small amount of overall spending it was almost 60 percent of the total amount of federal purchases.
For the purposes of this article, I am ignoring issues about conflating government purchases and private sector purchases. Government spending is a burden on the economy, and equating government purchases with voluntary private purchases is at best questionable. For more on this issue, I recommend Murray Rothbard’s explanation in his America’s Great Depression that government taxes and spending are “depredations” on the economy.
We now get to net exports. This component of GDP appears to create the most confusion. Some critics of the use of GDP as a statistical estimate of US production conclude that the net exports category results in a statistic that understates our annual production. Let me try to clear up this issue.
Net exports equal exports minus imports. For 2019, we purchased $3,135.7 billion of imports and sold $2,503.8 billion of exports, leading to a trade deficit of $632 billion. Net exports were negative and were subtracted from the other categories of expenditures.
Exports are goods and services produced in the US and sold abroad, and of course this production should be included in GDP. However, it is not immediately obvious why they subtract imports from exports when calculating net exports.
When we purchase imports, those purchases are added to GDP. If we purchase an imported truck as a consumer good, that spending is added to personal consumption, if the government purchases the truck, it is added to government purchases; if a business buys the truck, that spending is added to investment; and if the truck is then exported, the dollar value of the truck is added to exports. Foreign production has nothing to do with our estimate of our gross domestic production. In order to ignore foreign production that is imported to the US, they subtract imports in the net exports category.
Imports are added into GDP when we purchase them. Therefore, it is entirely appropriate to subtract imports when calculating net exports. In this manner, imports do not affect the calculation of GDP.
In explaining the components of gross domestic product, I have made the following major points:
- Subtracting decreases in inventories from GDP is legitimate, since the purchases of those inventories are added elsewhere into GDP.
- Intermediate goods are not a part of the GDP calculation unless those intermediate goods are added to inventories.
- Importantly, subtracting imports when calculating net exports makes perfect sense, since imports are added to other categories of GDP. By adding imports to GDP, then subtracting them out of GDP, imports have no effect on the GDP statistic.
I trust that I have made it clear that I am not defending gross domestic product as a sound measure of our country’s economic performance. There are many reasons we should be skeptical of this statistic. In this article, I am simply attempting to clarify the issues regarding the calculation of this statistic.
Gross Domestic Product, 2019
GDP (billions of dollars)
Gross Private Domestic Investment
Changes in Inventories
State and Local Purchases
Source: 2020 Economic Report of the President.
- 1. For those interested in an analysis of the validity of intellectual property rights, I recommend Butler Shaffer’s A Libertarian Critique of Intellectual Property (Auburn, AL: Mises Institute, 2013).