Go to the library. Pull Historical Statistics of the United States down off of the shelf. Perform a few calculations, and discover that GDP per worker in the United States today is some $55,000 dollars per year--measured at 1995's prices--and that GDP per worker in the United States a little over a century ago--back in 1890--was some $12,000 a year, when measured at 1995's prices. (When measured at 1890's prices, GDP per worker in 1890 was more like $500 a year: inflation has multiplied the price level some twenty five-fold since 1890. When measured at 1890's prices, real GDP per worker today is infinity--there are so many products, commodities, and services that we use today that could not have been obtained at any price a century ago.)
What Real GDP per Worker Is:
Let's unpack the chart title, because there is a lot going on in it: U.S. and 1890-1995 are relatively straightforward. GDP is an abbreviation for gross domestic product. "Gross" means that we are not correcting for depreciation--the reduction in value of economic capital as it slowly wears out and approaches the end of its useful life. Four years ago my wife and I got a Volvo station wagon costing some $22,000. Today the current value of this station wagon is less than $11,000--four years' worth of wear-and-tear that have brought it four years closer to the end of its useful life have also reduced its economic value by half.
Measures of net economic product correct for depreciation--calculate, for example, that a factory that has total gross profits of $3,000,000 during a year didn't really make $3,000,000 but instead only $1,000,000, because the process of production put $2,000,000 worth of wear-and-tear on the factory's capital stock--wear-and-tear that will require $2,000,000 worth of maintenance to restore the factory to its roughly original condition.
Measures of gross economic product do not correct for depreciation. Measures of net product are, conceptually, better--but they are very hard to do, the people at the Commerce Department's Bureau of Economic Analysis who create these numbers have little confidence in the accuracy of their depreciation estimates, and they prefer to focus on numbers they think they can measure adequately--even if it is not quite the concept they would ideally like to measure.
Domestic means that we are looking at all marketed and government production taking place inside the boundaries of the United States. We don't care that New York's Rockefeller Center is owned by people who live in Japan--the services provided by Rockefeller Center to those who rent office space there, and the income generated by the rental are part of domestic product. Conversely, income generated abroad by factories located in Malaysia owned by U.S. citizens does not enter into domestic product.
The alternative concept, "national" product, would include products made and income generated by extra-U.S. property owned by U.S. citizens, and would exclude products made and income generated by property in the U.S. that was owned by non-U.S. citizens. We now use domestic product because our estimates of cross-border income and profit flows are riddled with error, and thus our estimates of national product are of signficantly lower quality than our estimates of domestic product.
Product. A noun. The economic product of a country, a region, an individual is the market value of goods and services produced over the course of a year. For example, my mother the psychologist "produces" some $80,000 of therapy each year. She works. Patients feel better and more well-adjusted--and find it worthwhile to pay her--or their insurance companies do. That $80,000 is her income, and it is also value-added for the economy as a whole: something produced over the course of a year which consumers were willing to pay for.
Real and (1995 Prices). Measured economic product could change because the volume of economic activity changed, or it could change because the prices at which goods and services sell changed--either because of general inflation or deflation, or because of shifts in relative prices. We want to ignore shifts in measured economic product caused by shifts in the price level. So we look at real GDP at 1995 prices. The idea is to take a representative slice of what was produced at some other date, and ask "what would this sell for if we brought it forward in time to 1995?"
Per Worker. Real GDP is a measure only of economic activity that passes through the market--is bought or sold (with a few exceptions). Within-the-household-production is counted in GDP if it is bought or paid for, and if not, is not. As the share of the American adult population in the paid labor force has risen, so measured GDP has risen even though part of what has been going on has been the shifting boundary between categories of work that used to be outside, but are now inside the market. So we divide real GDP by the size of the American labor force to attempt to control for the shifting boundary between market and non-market work, and also to control for the overall growth of population.
So there is the unpacked title of the chart. It is a measure of the average
productivity, controlling for inflationary and deflationary shifts in the
price level, of the American labor force.
"Average" in the sense that we have taken the market value of
all goods and services produced in the U.S. and divided by the number of
workers. It is a gross measure in that it doesn't take account of
depreciation and capital consumption--the fact that this year's production
has placed wear-and-tear on the nation's accumulated capital stock. When
we look at this graph, what do we see? It has gone up a lot over the past
century. In 1890, real GDP per worker (at 1995's prices) was only some $12,000
a year. Take what the average worker produced in 1890, bring it forward
in time to 1995, and sell it--and you will get some $12,000 for it. By contrast,
real GDP per worker crossed $55,000 a year last year, and continues to rise.
Historical Statistics Gives Us an Underestimate of Growth:
In fact, the factor of 4.5 is almost surely a major underestimate of the increase in our material prosperity over the past century for two reasons. First, but less important, the work year has shrunk considerably over the past century. We no longer see twelve-hour shifts in steel mills. Total--measured--product per hour worked is more like six times what it was in 1890.
Second, and more important, the question answered by the graph above--"how much would 1890s production be worth if we stuffed it in a time machine, brought it forward in time to today, and then sold it on the open market?"--is not really the right question to ask if we want to compare material standards of living.
Why is it the wrong question? Consider a thought experiment. My family's income today is roughly $110,000 a year, about twice average GDP per worker. The Historical Statistics tell us that we today have some six times the GDP per work hour of 1890. So suppose that you stuff me and my family into a time machine, send us back a century to 1890, and then give us an income equal to twelve times that of 1890's average GDP per worker.
Does that give me and my family--back in 1890--roughly the same material standard of living that we have today in 1996? Could we buy commodities that would make us roughly as happy, from a material welfare standpoint, then as we are now?
The answer is surely not. The first thing I want to buy is health insurance: the ability to go to the doctor and be treated with late-twentieth-century medicines. The second thing I want to buy is a house with electrical and gas utility hookups, central heating, and the standard portfolio of appliances. The third thing I want to buy is access to information--audio and video broadcasts, recorded music, computing power and access to databases.
None of these were available back in 1890.
I could not buy a washing machine; instead I am offered the opportunity to hire a servant to do the household's washing. I can't buy modern medical technology; instead I am offered an array of remedies that promise a fifty-fifty chance of making me sicker. I can't buy airplane flights; instead I am offered long distance train and boat tickets that churn up enormous amounts of time in travel.
A large chunk of my--high--material standard of living today is not only the amount of purchasing power that my income gives me, but also the very broad range of commodities that I can choose to purchase. When we say that average GDP per worker in 1890 was equal to some $12,000 at 1995 prices, we cannot help but think that the material standard of living then was about what we could obtain now if we had $12,000 to spend. But that leaves out a very important part of the picture: the material standard of living then was about what we could obtain now if we had $12,000 to spend but were required to spend it all on commodities that have been around for more than a century: no modern entertainment or communications or transportation technologies; no modern appliances; buildings, roads, bridges, and other infrastructure built using century-old technologies.
How much unrestricted money would you think of as a fair trade for your $12,000 1995-price dollars that could be spent only on last century's goods? There is no way to arrive at an "objective" measure independent of human psychology. And, indeed, human psychology is shaped by society and technology--perhaps people of a century ago would place much lower value on the increased range of consumer choice made possible by modern technology than we do. In fact, the convention used by Historical Statistics in calculating long-run growth--stuff the past's output into a time machine, bring it forward to today, and sell it--was set in stone by Simon Kuznets because it was the most sensible objective measure of long-run economic growth he could think of, even though it did not answer the question that he really wanted to ask.
My own guess is that, if confined to purchasing and consuming only those commodities that were in the set of items producible in 1890, it would take at least $350,000 to make me as happy, in a material standard of living sense, as I am today. It might be impossible, there might be no amount of money that would do if confined to commodities producible using 1890's technology: I would probably, and at least one of my two children would be probably, dead without antibiotics to fight pneumonia and intravenous solutions to raise blood sugar levels.
So my own guess is that, if you have to pick a single number to summarize the explosion in America's material well-being over the past century, the best number to pick is not six but twenty: that we today have twenty times the productive capabilities and material wealth of our predecessors of a century ago.
Pre-Twentieth Century Economic Growth:
Economic growth in this twentieth century has been vastly greater than in any previous century. All other eras appear static compared to the twentieth century: changes in relative wealth of magnitudes that used to take centuries are compressed into decades.
The nineteenth century saw, according to Historical Statistics, perhaps a doubling of material standards of living in the United States--perhaps a tripling once proper account is taken of the expanded range of technological capabilities. The standard of living in the Netherlands--the richest economy in the world at the end of the eighteenth century--was perhaps fifty percent higher than it had been three centuries before, at the time of the Renaissance.
But before that? Between the invention of agriculture and the industrial revolution, wealth and technology developed slowly. Medieval historians speak of centuries and half-millennia when they speak of the pace of the diffusion of key inventions like the watermill, or the heavy plow, or the horse collar. Edward Gibbon, writing in the 18th century of the decline and fall of the Roman Empire, comes up with only a short list of differences between life in eighteenth century Britain and life in Imperial Rome. The two most striking that Gibbon notes are that Julius Cæsar's heir Augustus had no linen for his bed, and that Roman carts had no springs (while English carriages of what Gibbon and his contemporaries liked to call their "Augustan Age" did). Yet almost two millennia separate the Emperor Augustus from Edward Gibbon.
So slow was the pace of change that people, or at least aristocratic intellectuals, could think of their predecessors of a thousand years before or more as effectively their contemporaries. And they were not far wrong. Marcus Tullius Cicero, a Roman aristocrat and politician of the generation before Augustus, might well have felt more or less at home in the company of Virginia planter Thomas Jefferson. The slaves outside grew different crops. The ploughs were better in Jefferson's time. Sailing ships were much improved.
Printing technology would have struck Cicero as amazing and wonderful: for Cicero acquiring one copy of one book involved two months' worth of copying labor by a literate slave, an amount of labor that we would value at perhaps $4,000 dollars compared to the $10 price of a trade paperback book today; we today find the real price of books in terms of human labor to be 1/400 of what it was for Cicero, and even in Jefferson's day the real price of books had already fallen to perhaps 1/50 of what it had been at the beginning of the Roman Empire.
But differences in standards of living and in technologies used to manipulate the world were small enough that Cicero would have found the standard of living and style of life familiar. He would have thought his own material well-being perhaps quantitatively, but certainly not qualitatively, inferior to Jefferson's.
Even the first century of the industrial revolution produced "improvements" and not "revolutions" in standards of living. With the railroad and the spinning and weaving of textiles as very important exceptions, most innovations during the first century or so of the industrial revolution proper were innovations in transportation and in production processes and types of capital goods. Standards of living improved because of these innovations in production processes and capital goods. But styles of life remained much the same, for improvements in productivity in the first half of the nineteenth century were concentrated in a few relatively narrow sectors rather than spread throughout the economy.
So slow was the pace of improvement that, while the literary intellectuals who developed economics in the first half of the 19th century were aware that they were living through a hitherto unseen economic transformation, they debated whether this industrial revolution was an improvement or a degeneration. While economists were unsure, non-economists were predominantly of the belief that the rich were getting richer and the poor poorer in Britain even as it was undergoing the industrial revolution.
Rough estimates of relative changes in material standards of living across the centuries are extremely hazardous. But the figure below does not do violence to the qualitative picture as it tries to indicate the relative economic growth over each of the past ten centuries of the leading-edge economies.
Twentieth Century Growth: Details
Computers, automobiles, airplanes, VCR's, washing machines, vacuum cleaners, telephones, and other technologies--combined with mass production--give middle-class citizens of the United States degrees of material wealth-control over commodities, and the ability to consume services-that previous generations could never have imagined. In fact, it is hard for even us to imagine what such an enormous gulf--sixfold at a minimum, more like twenty-fold in my estimation, vastly larger in the opinion of others--could mean.
One way to get a handle on such rapid growth is to try to take a look at its details. One gauge comes from examining the Montgomery Ward catalog. At the turn of the century Montgomery Ward was the largest mail-order business in the United States, supplying rural and small-town households around the country with goods produced in America's factories. Montgomery Ward was big business: fully half of Americans then still lived on the farm or in very small towns. The shops and stores of the big cities were much less convenient than the regular arrival of the mail-order catalogues. Shipping by mail order from centralized warehouses, companies like Montgomery Ward were willing to supply goods ranging from sterling silver teaspoons to sets of the Encyclopedia Brittanica to drill presses.
[]
Send e-mail to Brad DeLong at delong@econ.berkeley.edu