Bruce Bartlett held senior policy roles in the Reagan and George H.W. Bush administrations and served on the staffs of Representatives Jack Kemp and Ron Paul. He is the author of “The Benefit and the Burden: Tax Reform â€" Why We Need It and What It Will Take.â€
My last post, on the impact of federal stimulus on economic growth, appears to have confused some readers. My purpose was to explain the basics of national income accounting and how stimulus fit in. The point was that some federal spending is stimulative by definition. Other spending may or may not be stimulative; only analysis can determine that.
Let me revisit national income accounts to note that they are like double-entry bookkeeping: two completely different things, debits and credits, must necessarily add up. Because they are calculated separately from different sources, the fact that they must add up allows one to be a check on the accuracy of the ot her.
In my previous post, I explained that the gross domestic product is calculated by adding together consumption, investment, net exports (exports of goods and services minus imports) and government consumption or investment at all levels (federal, state and local).
But one can just as well look at G.D.P. in two other ways. The first is to decompose G.D.P. into a different set of categories consisting of sales of goods, services and structures, whether for consumption or investment, to their final users, whether they are businesses or individuals.
Final sales of goods and services is basically the same as G.D.P. except for changes in inventories. When businesses add to inventories, it raises G.D.P.; when they draw down inventories, it subtracts from G.D.P., which is intended to measure current production. Inventories represent past production.
Within the category of final sales are three major subcategories: goods, services and structures (buildings, houses). Within goods are durable and nondurable goods. Nondurable goods are items like food and clothing that are consumed relatively quickly or used up. Durable goods are longer-lived, such as automobiles, computers and industrial equipment that will last for years.
In 2011, nominal G.D.P. was $15,094 billion. Of this, purchases of goods accounted for $4,259 billion, or 28 percent. Of the goods, about half were durable ($2,186 billion) and half nondurable ($2,073 billion). Consumption of services accounted for $9,812 billion of G.D.P. or almost two-thirds. Production of structures added $1,022 billion or 7 percent.
Another way of looking at G.D.P. is to add up various types of income. This is done in a separate statistic called gross domestic income. In theory, it should equal G.D.P. and thus is a check on its accuracy, just as debits and credits are.
The following table shows how gross domestic income is calculated. N ote that although it should equal G.D.P., it is in fact slightly higher. This may change through future data revisions.
Some economists believe that gross domestic income may be a better measure of the economy for analytical purposes. One problem is that the underlying data necessary to calculate the income side of the economy comes largely from tax records, which tend to be slower in being available than data from industrial production and other components of G.D.P.
Some readers may be wondering where government transfer payments are in the national income and product accounts. They don't appear in G.D.I., because transfers don't arise from current economic activity, as is the case with wages and profits. Note that certain taxes are a positive figure in G.D.I. to account for the fact that they are deductible by businesses in calculating profits.
Transfers can be found in a separate calculation for personal income, which consists of compensation for workers , proprietors' income, rental income, interest and dividend income, and transfers. However, social insurance taxes are subtracted to avoid double-counting.
Obviously, all this is very complicated, and economists are constantly debating theoretical questions about how G.D.P. and other key economic figures are calculated. These issues are frequently discussed in a government publication called the Survey of Current Business, available from the Bureau of Economic Analysis.
One hotly debated issue in recent years has been accounting for environmental degradation in the national income data. In 1999, the National Research Council published a report urging a number of changes that would improve our understanding of mineral extraction, renewable energy and related issues.
Another issue is the extent to which figures such as G.D.P. relate to happiness and well-being. In 2009, the economists Joseph E. Stiglitz and Amartya Sen, both winners of the Nobel Prize, publis hed a study pointing to a number of ways in which measures of economic growth and well-being are in conflict.
The Organization for Economic Cooperation and Development now publishes data comparing major countries on a wide variety of measures of well-being beyond those narrowly economic in nature, such as community, civic engagement, life satisfaction, safety and work-life balance.
Whether a particular set of data is appropriate or appropriately calculated really depends on the question being asked. While G.D.P. is often used as a summary measure of national wealth, that was never really its purpose. Policy makers were mainly interested in economic changes â€" how well the economy as a whole was doing, quantitatively, this year relative to last year or previous years. They were also interested in timely data so that policies could be adjusted.
One problem with making major conceptual changes to the calculation of G.D.P. and other macroeconomic data is that we have many decades of economic research correlating it with monetary policy, tax policy and a wide variety of other factors. Unless a conceptual change permits creation of a consistent historical time series, it could end up creating a lot of confusion that will do more harm than good.
