Gross domestic product, or GDP, is often considered the best measure of how well an economy is performing. In the United States, this statistic is computed every three months by the Bureau of Economic Analysis, a part of the U.S. Department of Commerce, from a large number of primary data sources. These primary sources include both (1) administrative data, which are byproducts of government functions such as tax collection, education programs, defense, and regulation, and (2) statistical data, which come from government surveys of, for example, retail establishments, manufacturing firms, and farms. The purpose of GDP is to summarize all these data with a single number representing the dollar value of economic activity in a given period of time.

There are two ways to view this statistic. One way to view GDP is as the total income of everyone in the economy; another way is as the total expenditure on the economy’s output of goods and services. From either viewpoint, it is clear why GDP is a gauge of economic performance. GDP measures something people care about—their incomes. Similarly, an economy with a large output of goods and services can better satisfy the demands of households, firms, and the government.

How can GDP measure both the economy’s income and its expenditure on output? The reason is that these two quantities are really the same: for the economy as a whole, income must equal expenditure. That fact, in turn, follows from an even more fundamental one: because every transaction has a buyer and a seller, every dollar of expenditure by a buyer must become a dollar of income to a seller. When Jack paints Jill’s house for $10,000, that $10,000 is income to Jack and expenditure by Jill. The transaction contributes $10,000 to GDP, regardless of whether we are adding up all income or all expenditure.

To understand the meaning of GDP more fully, we turn to national income accounting, the system used to measure GDP and many related statistics.

Imagine an economy that produces a single good, bread, from a single input, labor. Figure 2-1 illustrates all the economic transactions that occur between households and firms in this economy.

19

The inner loop in Figure 2-1 represents the flows of bread and labor. The households sell their labor to the firms. The firms use the labor of their workers to produce bread, which the firms in turn sell to the households. Hence, labor flows from households to firms, and bread flows from firms to households.

The outer loop in Figure 2-1 represents the corresponding flow of dollars. The households buy bread from the firms. The firms use some of the revenue from these sales to pay the wages of their workers, and the remainder is the profit belonging to the owners of the firms (who themselves are part of the household sector). Hence, expenditure on bread flows from households to firms, and income in the form of wages and profit flows from firms to households.

GDP measures the flow of dollars in this economy. We can compute it in two ways. GDP is the total income from the production of bread, which equals the sum of wages and profit—the top half of the circular flow of dollars. GDP is also the total expenditure on purchases of bread—the bottom half of the circular flow of dollars. To compute GDP, we can look at either the flow of dollars from firms to households or the flow of dollars from households to firms.

20

These two ways of computing GDP must be equal because, by the rules of accounting, the expenditure of buyers on products is income to the sellers of those products. Every transaction that affects expenditure must affect income, and every transaction that affects income must affect expenditure. For example, suppose that a firm produces and sells one more loaf of bread to a household. Clearly this transaction raises total expenditure on bread, but it also has an equal effect on total income. If the firm produces the extra loaf without hiring any more labor (such as by making the production process more efficient), then profit increases. If the firm produces the extra loaf by hiring more labor, then wages increase. In both cases, expenditure and income increase equally.

21

In an economy that produces only bread, we can compute GDP by adding up the total expenditure on bread. Real economies, however, include the production and sale of a vast number of goods and services. To compute GDP for such a complex economy, it will be helpful to have a more precise definition: Gross domestic product (GDP) is the market value of all final goods and services produced within an economy in a given period of time. To see how this definition is applied, let’s discuss some of the rules that economists follow in constructing this statistic.

Adding Apples and Oranges The U.S. economy produces many different goods and services—hamburgers, haircuts, cars, computers, and so on. GDP combines the value of these goods and services into a single measure. The diversity of products in the economy complicates the calculation of GDP because different products have different values.

Suppose, for example, that the economy produces four apples and three oranges. How do we compute GDP? We could simply add apples and oranges and conclude that GDP equals seven pieces of fruit. But this makes sense only if we think apples and oranges have equal value, which is generally not true. (This would be even clearer if the economy produces four watermelons and three grapes.)

To compute the total value of different goods and services, the national income accounts use market prices because these prices reflect how much people are willing to pay for a good or service. Thus, if apples cost $0.50 each and oranges cost $1.00 each, GDP would be

GDP equals $5.00—the value of all the apples, $2.00, plus the value of all the oranges, $3.00.

Used Goods When the Topps Company makes a pack of baseball cards and sells it for $2, that $2 is added to the nation’s GDP. But when a collector sells a rare Mickey Mantle card to another collector for $500, that $500 is not part of GDP. GDP measures the value of currently produced goods and services. The sale of the Mickey Mantle card reflects the transfer of an asset, not an addition to the economy’s income. Thus, the sale of used goods is not included as part of GDP.

The Treatment of Inventories Imagine that a bakery hires workers to produce more bread, pays their wages, and then fails to sell the additional bread. How does this transaction affect GDP?

The answer depends on what happens to the unsold bread. Let’s first suppose that the bread spoils. In this case, the firm has paid more in wages but has not received any additional revenue, so the firm’s profit is reduced by the amount that wages have increased. Total expenditure in the economy hasn’t changed because no one buys the bread. Total income hasn’t changed either—although more is distributed as wages and less as profit. Because the transaction affects neither expenditure nor income, it does not alter GDP.

22

Now suppose, instead, that the bread is put into inventory (perhaps as frozen dough) to be sold later. In this case, the national income accounts treat the transaction differently. The owners of the firm are assumed to have “purchased” the bread for the firm’s inventory, and the firm’s profit is not reduced by the additional wages it has paid. Because the higher wages paid to the firm’s workers raise total income, and the greater spending by the firm’s owners on inventory raises total expenditure, the economy’s GDP rises.

What happens later when the firm sells the bread out of inventory? This case is much like the sale of a used good. There is spending by bread consumers, but there is inventory disinvestment by the firm. This negative spending by the firm offsets the positive spending by consumers, so the sale out of inventory does not affect GDP.

The general rule is that when a firm increases its inventory of goods, this investment in inventory is counted as an expenditure by the firm owners. Thus, production for inventory increases GDP just as much as does production for final sale. A sale out of inventory, however, is a combination of positive spending (the purchase) and negative spending (inventory disinvestment), so it does not influence GDP. This treatment of inventories ensures that GDP reflects the economy’s current production of goods and services.

Intermediate Goods and Value Added Many goods are produced in stages: raw materials are processed into intermediate goods by one firm and then sold to another firm for final processing. How should we treat such products when computing GDP? For example, suppose a cattle rancher sells one-quarter pound of meat to McDonald’s for $1, and then McDonald’s sells you a hamburger for $3. Should GDP include both the meat and the hamburger (a total of $4) or just the hamburger ($3)?

The answer is that GDP includes only the value of final goods. Thus, the hamburger is included in GDP but the meat is not: GDP increases by $3, not by $4. The reason is that the value of intermediate goods is already included as part of the market price of the final goods in which they are used. To add the intermediate goods to the final goods would be double counting—that is, the meat would be counted twice. Hence, GDP is the total value of final goods and services produced.

One way to compute the value of all final goods and services is to sum the value added at each stage of production. The value added of a firm equals the value of the firm’s output less the value of the intermediate goods that the firm purchases. In the case of the hamburger, the value added of the rancher is $1 (assuming that the rancher bought no intermediate goods), and the value added of McDonald’s is $3 – $1, or $2. Total value added is $1 + $2, which equals $3. For the economy as a whole, the sum of all value added must equal the value of all final goods and services. Hence, GDP is also the total value added of all firms in the economy.

Housing Services and Other Imputations Although most goods and services are valued at their market prices when computing GDP, some are not sold in the marketplace and therefore do not have market prices. If GDP is to include the value of these goods and services, we must use an estimate of their value. Such an estimate is called an imputed value.

23

Imputations are especially important for determining the value of housing. A person who rents a house is buying housing services and providing income for the landlord; the rent is part of GDP, both as expenditure by the renter and as income for the landlord. Many people, however, own their homes. Although they do not pay rent to a landlord, they are enjoying housing services similar to those that renters purchase. To take account of the housing services enjoyed by homeowners, GDP includes the “rent” that these homeowners “pay” to themselves. Of course, homeowners do not in fact pay themselves this rent. The Department of Commerce estimates what the market rent for a house would be if it were rented and includes that imputed rent as part of GDP. This imputed rent is included both in the homeowner’s expenditure and in the homeowner’s income.

Imputations also arise in valuing government services. For example, police officers, firefighters, and senators provide services to the public. Assigning a value to these services is difficult because they are not sold in a marketplace and therefore do not have a market price. The national income accounts include these services in GDP by valuing them at their cost. That is, the wages of these public servants are used as a measure of the value of their output.

In many cases, an imputation is called for in principle but, to keep things simple, is not made in practice. Because GDP includes the imputed rent on owner-occupied houses, one might expect it also to include the imputed rent on cars, lawn mowers, jewelry, and other durable goods owned by households. Yet the value of these rental services is left out of GDP. In addition, some of the output of the economy is produced and consumed at home and never enters the marketplace. For example, meals cooked at home are similar to meals cooked at a restaurant, yet the value added when a person prepares a meal at home is left out of GDP.

Finally, no imputation is made for the value of goods and services sold in the underground economy. The underground economy is the part of the economy that people hide from the government either because they wish to evade taxation or because the activity is illegal. Examples include domestic workers paid “off the books” and the illegal drug trade. The size of the underground economy varies widely from country to country. In the United States, the underground economy is estimated to be less than 10 percent of the official economy, whereas in some developing nations, such as Thailand, Nigeria, and Bolivia, the underground economy is more than half as large as the official one.

Because the imputations necessary for computing GDP are only approximate, and because the value of many goods and services is left out altogether, GDP is an imperfect measure of economic activity. These imperfections are most problematic when comparing standards of living across countries. Yet as long as the magnitude of these imperfections remains fairly constant over time, GDP is useful for comparing economic activity from year to year.

Economists use the rules just described to compute GDP, which values the economy’s total output of goods and services. But is GDP a good measure of economic well-being? Consider once again the economy that produces only apples and oranges. In this economy, GDP is the sum of the value of all the apples produced and the value of all the oranges produced. That is,

24

Economists call the value of goods and services measured at current prices nominal GDP. Notice that nominal GDP can increase either because prices rise or because quantities rise.

It is easy to see that GDP computed this way is not a good gauge of economic well-being. That is, this measure does not accurately reflect how well the economy can satisfy the demands of households, firms, and the government. If all prices doubled without any change in quantities, nominal GDP would double. Yet it would be misleading to say that the economy’s ability to satisfy demands has doubled because the quantity of every good produced remains the same.

A better measure of economic well-being would tally the economy’s output of goods and services without being influenced by changes in prices. For this purpose, economists use real GDP, which is the value of goods and services measured using a constant set of prices. That is, real GDP shows what would have happened to expenditure on output if quantities had changed but prices had not.

To see how real GDP is computed, imagine we want to compare output in 2014 with output in subsequent years for our apple-and-orange economy. We could begin by choosing a set of prices, called base-year prices, such as the prices that prevailed in 2014. Goods and services are then added up using these base-year prices to value the different goods in each year. Real GDP for 2014 would be

Similarly, real GDP in 2015 would be

And real GDP in 2016 would be

Notice that 2014 prices are used to compute real GDP for all three years. Because the prices are held constant, real GDP varies from year to year only if the quantities produced vary. Because a society’s ability to provide economic satisfaction for its members ultimately depends on the quantities of goods and services produced, real GDP provides a better measure of economic well-being than does nominal GDP.

25

From nominal GDP and real GDP we can compute a third statistic: the GDP deflator. The GDP deflator, also called the implicit price deflator for GDP, is the ratio of nominal GDP to real GDP:

The GDP deflator reflects what’s happening to the overall level of prices in the economy.

To better understand this, consider again an economy with only one good, bread. If P is the price of bread and Q is the quantity sold, then nominal GDP is the total number of dollars spent on bread in that year, P × Q. Real GDP is the number of loaves of bread produced in that year times the price of bread in some base year, P_base × Q. The GDP deflator is the price of bread in that year relative to the price of bread in the base year, P/P_base.

The definition of the GDP deflator allows us to separate nominal GDP into two parts: one part measures quantities (real GDP) and the other measures prices (the GDP deflator). That is,

Nominal GDP = Real GDP × GDP Deflator.

Nominal GDP measures the current dollar value of the output of the economy. Real GDP measures output valued at constant prices. The GDP deflator measures the price of output relative to its price in the base year. We can also write this equation as

In this form, you can see how the deflator earns its name: it is used to deflate (that is, take inflation out of) nominal GDP to yield real GDP.

We have been discussing real GDP as if the prices used to compute this measure never change from their base-year values. If this were truly the case, over time the prices would become more and more dated. For instance, the price of computers has fallen substantially in recent years, while the price of a year at college has risen. When valuing the production of computers and education, it would be misleading to use the prices that prevailed ten or twenty years ago.

To solve this problem, the Bureau of Economic Analysis used to periodically update the prices used to compute real GDP. About every five years, a new base year was chosen. The prices were then held fixed and used to measure year-to-year changes in the production of goods and services until the base year was updated once again.

In 1995, the Bureau announced a new policy for dealing with changes in the base year. In particular, it now uses chain-weighted measures of real GDP. With these new measures, the base year changes continuously over time. In essence, average prices in 2014 and 2015 are used to measure real growth from 2014 to 2015, average prices in 2015 and 2016 are used to measure real growth from 2015 to 2016, and so on. These various year-to-year growth rates are then put together to form a “chain” that can be used to compare the output of goods and services between any two dates.

26

This new chain-weighted measure of real GDP is better than the more traditional measure because it ensures that the prices used to compute real GDP are never far out of date. For most purposes, however, the differences are not significant. It turns out that the two measures of real GDP are highly correlated with each other. As a practical matter, both measures of real GDP reflect the same thing: economy-wide changes in the production of goods and services.

27

Economists and policymakers care not only about the economy’s total output of goods and services but also about the allocation of this output among alternative uses. The national income accounts divide GDP into four broad categories of spending:

Thus, letting Y stand for GDP,

Y = C + I + G + NX.

GDP is the sum of consumption, investment, government purchases, and net exports. Each dollar of GDP falls into one of these categories. This equation is an identity—an equation that must hold because of the way the variables are defined. It is called the national income accounts identity.

Consumption consists of household expenditures on goods and services. Goods are tangible items, and they in turn are divided into durables and nondurables. Durable goods are goods that last a long time, such as cars and TVs. Nondurable goods are goods that last only a short time, such as food and clothing. Services include various intangible items that consumers buy, such as haircuts and doctor visits.

Investment consists of items bought for future use. Investment is divided into three subcategories: business fixed investment, residential fixed investment, and inventory investment. Business fixed investment, also called nonresidential fixed investment, is the purchase by firms of new structures, equipment, and intellectual property products. (Intellectual property products include software, research and development, and entertainment, literary, and artistic originals.) Residential investment is the purchase of new housing by households and landlords. Inventory investment is the increase in firms’ inventories of goods (if inventories are falling, inventory investment is negative).

Government purchases are the goods and services bought by federal, state, and local governments. This category includes such items as military equipment, highways, and the services provided by government workers. It does not include transfer payments to individuals, such as Social Security and welfare. Because transfer payments reallocate existing income and are not made in exchange for goods and services, they are not part of GDP.

The last category, net exports, accounts for trade with other countries. Net exports are the value of goods and services sold to other countries (exports) minus the value of goods and services that foreigners sell us (imports). Net exports are positive when the value of our exports is greater than the value of our imports and negative when the value of our imports is greater than the value of our exports. Net exports represent the net expenditure from abroad on our goods and services, which provides income for domestic producers.

28

29

The national income accounts include other measures of income that differ slightly in definition from GDP. It is important to be aware of the various measures, because economists and the media often refer to them.

To see how the alternative measures of income relate to one another, we start with GDP and modify it in various ways. To obtain gross national product (GNP), we add to GDP receipts of factor income (wages, profit, and rent) from the rest of the world and subtract payments of factor income to the rest of the world:

GNP = GDP + Factor Payments from Abroad − Factor Payments to Abroad.

Whereas GDP measures the total income produced domestically, GNP measures the total income earned by nationals (residents of a nation). For instance, if a Japanese resident owns an apartment building in New York, the rental income he earns is part of U.S. GDP because it is earned in the United States. But because this rental income is a factor payment to abroad, it is not part of U.S. GNP. In the United States, factor payments from abroad and factor payments to abroad are similar in size—each representing about 4 percent of GDP—so GDP and GNP are quite close.

30

To obtain net national product (NNP), we subtract from GNP the depreciation of capital—the amount of the economy’s stock of plants, equipment, and residential structures that wears out during the year:

NNP = GNP − Depreciation.

In the national income accounts, depreciation is called the consumption of fixed capital. It equals about 16 percent of GNP. Because the depreciation of capital is a cost of producing the output of the economy, subtracting depreciation shows the net result of economic activity.

Net national product is approximately equal to another measure called national income. The two differ by a small correction called the statistical discrepancy, which arises because different data sources may not be completely consistent.

National Income = NNP − Statistical Discrepancy.

National income measures how much everyone in the economy has earned.

The national income accounts divide national income into six components, depending on who earns the income. The six categories, and the percentage of national income paid in each category in 2013, are the following:

A series of adjustments take us from national income to personal income, the amount of income that households and noncorporate businesses receive. Four of these adjustments are most important. First, we subtract taxes on production and imports because these taxes never enter anyone’s income. Second, we reduce national income by the amount that corporations earn but do not pay out, either because the corporations are retaining earnings or because they are paying taxes to the government. This adjustment is made by subtracting corporate profits (which equal the sum of corporate taxes, dividends, and retained earnings) and adding back dividends. Third, we increase national income by the net amount the government pays out in transfer payments. This adjustment equals government transfers to individuals minus social insurance contributions paid to the government. Fourth, we adjust national income to include the interest that households earn rather than the interest that businesses pay. This adjustment is made by adding personal interest income and subtracting net interest. (The difference between personal interest and net interest arises in part because interest on the government debt is part of the interest that households earn but is not part of the interest that businesses pay out.) Thus,

31

Next, if we subtract personal taxes, we obtain disposable personal income:

Disposable Personal Income = Personal Income − Personal Taxes.

We are interested in disposable personal income because it is the amount households and noncorporate businesses have available to spend after satisfying their tax obligations to the government.

Because real GDP and the other measures of income reflect how well the economy is performing, economists are interested in studying the quarter-to-quarter fluctuations in these variables. Yet when we start to do so, one fact leaps out: all these measures of income exhibit a regular seasonal pattern. The output of the economy rises during the year, reaching a peak in the fourth quarter (October, November, and December) and then falling in the first quarter (January, February, and March) of the next year. These regular seasonal changes are substantial. From the fourth quarter to the first quarter, real GDP falls on average about 8 percent.2

It is not surprising that real GDP follows a seasonal cycle. Some of these changes are attributable to changes in our ability to produce: for example, building homes is more difficult during the cold weather of winter than during other seasons. In addition, people have seasonal tastes: they have preferred times for such activities as vacations and Christmas shopping.

32

When economists study fluctuations in real GDP and other economic variables, they often want to eliminate the portion of fluctuations due to predictable seasonal changes. You will find that most of the economic statistics reported are seasonally adjusted. This means that the data have been adjusted to remove the regular seasonal fluctuations. (The precise statistical procedures used are too elaborate to discuss here, but in essence they involve subtracting those changes in income that are predictable just from the change in season.) Therefore, when you observe a rise or fall in real GDP or any other data series, you must look beyond the seasonal cycle for the explanation.3

34