Part III. Growth, Quality, Happiness, and the Poor
Modern Growth was a Factor of at Least Sixteen
The heart of the matter is, to fix ideas, sixteen. Real income per head nowadays exceeds that around 1700 or 1800 in, say, Britain and in other countries that have experienced modern economic growth by such a large factor as sixteen, at least.1 You, oh average participant in the British economy, go through at least sixteen times more food and clothing and housing and education per person in a day than your ancestors did two or three centuries ago. You in the American or the South Korean economy, compared to the wretchedness of former Smiths in 1653 or Kims in 1953, have done even better. And if one accounts at their proper value such novelties as jet travel and vitamin pills and instant messaging, then the factor of material improvement climbs even higher than sixteen-to eighteen, or thirty, or far beyond. No previous episode of enrichment for the average person approaches it, not the China of the Song Dynasty or the Egypt of the New Kingdom, not the glory of Greece or the grandeur of Rome.
No competent economist, regardless of her politics, denies the Great Fact. The economist Stephen Marglin, for example, emphasizes community, which he believes was undermined by the Fact and its accompanying rhetoric of prudence-only. As a convinced socialist he thinks that power and striving had more to do with the Fact than a free-market economist does.2 But a neo-Marxist economist and a free-market economist both accept the great magnitude of the enrichment as a Fact. Likewise the economic historian Gregory Clark emphasizes a Darwinian struggle for eminence, which he believes explains the Fact. As a recently persuaded eugenicist he thinks that people are fated to be who they were born to be, which a true liberal finds ethically alarming, and anyway scientifically dubious.3 But a eugenic economist and a liberal economist both accept that the Fact broke the Malthusian curse.
Yet many non-economists or non-historians, in their politics left or right, are suspicious of innovation and hostile to markets, and remain unaware of the magnitude. They know something happened, of course, and that the vulgar bourgeois apologist will claim a “progress” of some sort, probably disputable and in any case deeply damaging to the poor or to a graceful life. But the non-economists and the non-historians (of whatever politics) have little idea of how very enriching the Fact has been of ordinary poor people. If you ask the regular readers of The Nation or of The National Review how much more material ease the average American had gained by the time of President Clinton as compared with President Monroe they will come up with a figure such as. . . go ahead: make a guess. . . perhaps, 200 percent or even 400 percent, maybe 800 percent—not, as is the case, 1700 percent, a factor of nearly 18, which is a lower bound on the American history.
The lack of precision in the estimates is worth the attention of specialists. But it is not important for the purpose here. The British or American or Japanese or South Korean increase could have been 8 or 10 or 35 times its level in 1700, rather than 16 or 18, and leave the heart of the matter undisturbed. People had always produced and consumed about $3 a day. By now they consume $30 a day if they are average denizens of the world, and $137 if Norwegians. The scientific fact established over the past fifty years by the labors of economists and economic historians is that modern economic growth has been astounding. Simply astounding. Imagine getting along on $3 a day in London or Seattle.
“Real national income per head” purports to measure what is earned by the average person in the nation as a whole, abstracting from merely monetary inflation. It measures the stuff per person we have—the pounds of bread or the number of haircuts, back and sides—not the mere dollars or yen. That’s why economists call it “real,” a word they favor. Thomas More disdained the grotesque consumerism of his early sixteenth-century England in which “four or five woolen cloaks and the same number of silk shirts are not enough for one [very well-off] person, and if he is bit fastidious not even ten will do.”4 Nowadays the merely average person in England has the equivalent of twenty or thirty. The fastidious boast hundreds.
If your ancestors lived in Finland the factor of real material improvement is more like 29, the average Finn in 1700 being only 60 percent better off in material terms than the average African at the time. Look at what happened to the average Norwegian. In 1700 the Netherlands was the most bourgeois and therefore the richest country in the world, 70 percent better off per capita than the soon-to-be United Kingdom. So if your ancestors lived in the Netherlands the modern improvement is only a factor of roughly 10. But it is measured, as all these figures are, in the cautious way that does not take account of the high qualities of modern pills and housing and message-sending. The actual Dutch factor must be a great deal higher. In Japan the factor since 1700 is fully 35. 5 In South Korea the cautiously measured factor since 1953, when income per head despite access to some modern technology (motor trucks, electric lights) was about what it had been in Europe 450 years before, is almost 18. The South Korean revolution was crammed into four decades instead of, as in the first and British case, stretched out over three centuries.
Like the realization in astronomy during the 1920s that most of the “nebulae” detected by telescopes are in fact other galaxies unspeakably far from ours, the Great Fact of economic growth, discovered by historians and economists in the 1950s and elaborated since then, changes everything.
And in truth the amount by which average welfare multiplied under actually existing innovation exceeds by far the official and cautious statistics. Stuff unimaginable in 1700 or 1820 crowds our lives, from air conditioning to anesthesia. The new stuff makes the factors of 16 or 18 or even 30 gross understatements. William Nordhaus, a very useful economist at Yale, starts his paper on the economic history of lighting with the conventionally measured factor of 18 in American real income per head since 1800, or a factor of 13 if one is talking about real wages rather than real total income.6 But he notes what is known to all us expert economists (you amateurs will have to rely on common sense)—that the price indexes that are employed to take out the effects of inflation rise too steeply, because the stuff being priced gets better and gives more services for each supposedly inflation-corrected dollar. Air-conditioning instead of fans. Three-car garages in the standard house instead of one-car. Electric lights instead of candles.
It has happened recently, for example from 1970 to 1992, when the United States and many other countries saw a stagnation of real wages officially measured—the money wage divided by the official consumer price index. You will hear critics on the left saying that the ordinary person in the United States did not gain after 1970. They want to think that the Final Crisis of Capitalism is upon us. The leftward critics are not entirely wrong in their worry. But from 1970 to 1992 the conventional measure of prices didn’t adequately reflect the rising space per dollar’s worth of housing and the cheapening air-conditioning and the rarely puncturing automobile tires. Most economists reckon that on account of quality improvements the inflation rate conventionally measured was overstated in the period by about 1 percent a year (and continues to be to about the same extent).7 When allowing for the better quality of goods and services, therefore, the period of nominal stagnation in real wages witnessed a rise of about a third in the properly corrected real wage, which is what matters (together with health insurance supporting the Cadillac level of medical interventions that Americans insist on, which the talk of stagnating wages also doesn’t include—allowing for example immediate access to by-pass surgery that wasn’t used until the 1970s, and organ transplants even for some poor people, and none of the queuing for ordinary procedures that most other national systems have).8
A gain per head of merely 1 percent a year is not wonderful economic growth. The American average since 1820 has been more like 2 percent.9 Something bad did happen to the rate of innovation in the American economy 1970-1992, and the wages of ordinary folk did not rise at the rate they had 1945-1970. The event certainly bears examining, and lamenting. The economist Benjamin Friedman has shown how politics deteriorates as rates of growth decline towards zero.10 One percent is perilously close to zero, and sure enough the politics of the United States and other countries such as Britain in the period became correspondingly nasty. But neither was the growth among ordinary people literally zero, as the left so confidently and indignantly claims. Capitalism wasn’t in crisis 1970-1992. During and after those years it raised the standard of living of poor people worldwide at the fastest rate in history (and by the way, according to the economist Robert Gordon, after the dot-com boom the American economy stopped rewarding the very rich disproportionately). The real welfare of workers in the United States 1970-1992 did not in fact stagnate—as you can see in the statistics of housing space per person or automobiles per person or restaurant meals per person. It modestly rose, from continuing innovation. Anyone who lived through the period knows that it did, though the official and uncorrected statistics can overcome her common sense.
And the poor got much better off materially, even in the recent period of growing inequality. Robert Fogel’s point in his 2002 book is that the United States has a much smaller problem by now with the physical condition of the poor—this in contrast to 1900—than what he calls their “spiritual” condition.11 Michael Cox and Richard Alm made some controversial assertions in a book of 1999 about the class mobility of the American poor. But their statistics on what the poor consume are not controversial. They conclude that “Poor households of the 1990s in many cases compared favorably with an average family in owning the trappings of middle class life. For example, almost half the poor households in 1994 had air conditioners, compared to less than a third of the country as a whole in 1971.”12 That’s right, as anyone knows who lived during the 1970s and knew poor people, or was poor. During the 1940s, which some of us also lived through, really poor people in the American 1940s didn’t have running water or electricity or access to penicillin, and the merely average poor person didn’t have an automobile and lived in half the space that a poor person lives in now. In 1938 Americans had a car for every 4.4 people, in 1960 for every 2.4 people, in 2003 for every 1.26 of a person.
But the bigger, longer-term point is that correctly measuring the prices of things greatly increases the estimate of modern economic growth, 1800 to the present. Cox and Alm observe that a three-minute long-distance call across the U.S.A. in 1915 cost 90 hours of common labor.13 In 1999 it cost a minute and a half. No wonder your granny is always saying “This call must be costing you a fortune.” It once did. In 1900, Cox and Alm note, a pair of scissors cost the modern per-labor-hour equivalent of $67, which is why in the old days a middle-class Mother had the one pair, carefully guarded, and used it to make clothing, and only on special rainy days would she let Sis use it to cut up the old Sears catalogue for paper dolls. Fogel calculates that in 1875 in the United States the average family spent 74 percent of its income on food, clothing, and shelter. In 1995 it spent 13 percent.14
Nordhaus makes the point about the real cost of goods and services by studying over centuries the cost of one item, lighting.15 Illumination is easy to measure, in lumen hours per dollar of expenditure, say, or more to the point in the lumen hours per hour of human work to get the dollars. Conventional price indexes of lighting can be measured year-by-year with the money price of, say, candles for a while in the early nineteenth century, when they were the main source of indoor lighting. But between 1800 and 1992 it would be crazy to take the price of candles (used nowadays of course only for ceremonial purposes) as “the price” of lighting. No, the service of lighting, Nordhaus observes, became much cheaper in the nineteenth century with the marketing of whale oil, and then a lot cheaper again with kerosene, and then a whole lot cheaper with electric lighting, which itself has continued to cheapen down to the fluorescent replacements for incandescent bulbs we are now beginning to use. Cheap LED lighting cannot be far behind. In other words, we can easily follow the price of each such form of lighting in its own era, but not well across eras. The problem is worse for many products less measurable than lighting. What’s the early nineteenth century price of penicillin? Movies on TV? The Internet? How much would you pay in 1850 to get from Chicago to London in seven and a half hours?
We can, however, follow the candlepower per hour generated by lighting of various sorts in actual use and compare it to the labor hours required to buy it. Nordhaus confirms what you might expect if you’ve watched a lot of historical movies on TV: that the growth in effective lighting has been very large, measured in the tens of thousands of lumen hours per hour of labor. On South Dearborn Street in Chicago stands the 17-storey Monadnock Building, lovingly restored to its historical ambiance down to every visible detail. (Half of the Monadnock, finished in 1891, was the last Chicago skyscraper to depend on thick, load-bearing masonry; the southern half, started in 1891, was almost the first to depend on structural steel.) One of the restored details is the lighting in halls and elevators, with tiny incandescent lights reproducing the feeble glow of 1891. If you doubt that lighting has been revolutionized, visit the Monadnock Building.
Nordhaus reckons, to be roughly quantitative about it, that around 9000 B.C.E. it took 50 hours of labor to gather enough bundled sticks or whatever to achieve 1000 lumen hours of lighting (think of our ancestors deep in the Altamira caves drawing aurochs and horses and stick-figured humans hunting them). In 1800 with candles it took 5 hours (think of John Adams scribbling long letters to Talleyrand to prevent war with France). In 1900, thanks to kerosene and the new electric lights, feeble though they were, it took only 0.22 hours, a revolution. In 1992, thanks to the radical cheapening of electricity-based lighting, it took a mere 0.00012 hours. The outcome was a cheapening in eleven millennia by a factor of 417,000, and in the last two centuries alone by 41,700 (note the over-neat homology in the figures: Nordhaus is not claiming to measure very accurately; it is an order of magnitude he seeks). And the rate of fall in the past two centuries, of course, was immensely accelerated compared with the mere factor of 10 between the age of olive oil lamps in Roman times and the age of European candles in Georgian times—illustrating the stunning enrichment from very recent European technology. (And it casts a bright light, too, on the stunning Chinese exception as to the level of technology, if not its modern rate of change. In the fourth century B.C.E the Chinese were using natural gas for lighting, and later carried the gas about in bags.)16
Look around your house or street this evening and assess the lighting you get and how many tallow candles would be its equivalent—if you could cram in the candles, in the style of the Great Hall scenes in Harry Potter movies. If you fancy that it would be oh-so-romantic to live back in such ill-lit days, then the economic and social historians suggest gently that you think again. In the days of candles the average adult slept ten hours a night in winter rather than the eight he now sleeps. The miserably cold and dark house of an evening was literally not worth the candle.
Nordhaus extends the argument, more speculatively but plausibly, to other inventions such as airplanes, insulin, radar, telephones, and the rest, and in a rough guess to all sectors of the economy. (The great student of national income, Angus Maddison, scorns his calculations under a sneering heading: “Hallucinogenic History: Nordhaus and [Bradford] DeLong.” But in the passage Maddison stays, uncharacteristically, at the level of indignation, and gives no reasons.17 ) The cost of what an hour of work could buy of lighting and all sorts of things, Nordhaus reckons, has dramatically fallen since 1800 if you take into account the rise in the quality of categories such as “lighting” and “housing” and “transportation” and “medical care” and the rest.
Take medical care. The doctor and essayist Lewis Thomas, Dean of Yale’s and New York University’s medical schools, “the father of modern immunology,” believed that until the 1920s going to a doctor lowered your odds of survival. Most medical care was done at home, and a middle-class home in 1920 was always supplied with a big medical encyclopedia about how to care for scarlet fever and how to deliver babies. The biggest improvement didn’t come until the 1940s, with penicillin. Andrew Carnegie despite his wealth could not buy a cure for the pneumonia that killed his mother, a disease I myself have had twice, and was cured of the last time in three days.18 Or take psychiatry. Until the coming of psychotropic drugs, invented during the 1950s and in common clinical use by the 1970s, the psychiatrists had nothing to do for depression (and at one point for homosexuality) but to talk gently to you, and then in desperation apply electroshock.
Nordhaus concludes that from 1800 to 1992 in the American economy the real wage—the money wage divided by the prices of things, but properly corrected for their improving thingness—grew not by that conventionally and crudely measured factor of 13 but anywhere from a low estimate of a factor of 40 to a high of 190. One hundred and ninety. Good Lord. Call it as a rough and ready average a factor of 100. That’s one hundred times greater ability to buy with an hour of work. Two orders of magnitude.
If you run your eyes around your room now and try to push back in imagination to the life of your great-great-great-great-grandmother, you will find pretty reasonable a factor of 100 in per capita capacity-to-buy-the services-of-stuff. You are reading by a light many times brighter than the candlesticks your ancestor could bring to bear, and candles were anyway to be used sparingly, and only at dark of the moon, to get to the outhouse in Council Bluffs or to the end of a row in Salford without tripping and killing yourself. You by contrast have such light available in a score of places inside and outside your house. If you want to write to your lover it will be on a laptop with the calculating power of a building full of older “calculators” (until the 1940s the word meant “women employed to add up long columns of figures”), on which you can type effortlessly, and then e-mail the note to the other side of the world in a split second (instead of the gradually lengthening days or weeks the Postal Service requires). Or in scribbling a shopping list you can use a ball point pen which eases handwriting by a factor of perhaps six over quill and ink. You do not write much quicker, but you spend no time at all as your ancestor did sharpening quills or dipping ink—and the ink froze in the winter, because, remember, you have no central heating, and must write with gloves with those little holes at the tips of the fingers. And in any case the ball point with which you write, and the paper on which you write, costs a trivial amount of your time to buy, compared with earlier hours of work per fountain pen or paper sheet. When ball points were first introduced after World War II they were expensive like fountain pens, requiring many hours of your work to buy. Now you have 40 or 50 of them jammed in various coffee mugs around your house—by actual count I myself have about 100 (but after all my work is scribbling). The clerk in the store often forgets to take back his pen when you sign a credit-card slip. The credit facilities you enjoy are many times more efficient than the means of payment in 1800. The book you bought with the credit card costs a fraction of what a book did in 1800 in terms of human labor. The paper is cheap, the printing electronic, the binding is done by machine. Some bookstores now have automatic book-making machines with any of two million out-of-print titles available in twenty minutes. For this and thousands of other similar reasons your real income is vastly higher than that of your ancestors—and so you can have many more books than even Thomas Jefferson did, if you are a bookish sort, purchased with ease from Sandmeyer’s Books in Chicago or Powell’s Books in Portland or amazon.com in the ether. That is your widened scope. And on and on.
You can see the factor of 100 from the other, producing side of the economy in the frantic development of new and improved products for consumers. The economic historian Maxine Berg has argued persuasively for “incorporating product innovation [that is, new stuff] into the analysis of the industrial revolution.”19 She cites an American economic historian the late Kenneth Sokoloff arguing that new products drove a good deal of industrial innovation in the United States early in the nineteenth century, giving demand a role in innovation.20 Neglecting product innovation is what Nordhaus is complaining about: it results in a gigantic understatement of the rising scope of modern economies, because a light bulb (if you have electric service in your house, that is) is a much better consumer product than a candle. Against the focus on process innovation usual in studies of the Industrial Revolution, Berg finds in British patents in the eighteenth century an astonishing proliferation of carved or molded glass, retractable toast racks, japanning (with a polite bow to the reverse engineering of eastern inventions), tin plate buttons, and 115 patents for stamping, pressing, and embossing metals.21
Not that process innovations are to be set aside. But process innovation is itself entangled with product innovation. Berg notes that “producers of small tools as well as complex lathes and engines” that made for faster production of a given product “were often the same individuals producing ornamental stamped brassware, medallions and mechanical toys.”22 Products for consumers led to producers’ goods for factories. And the correct measurement of producers’ goods has the same problem of better and better quality that the measurement of consumers’ goods has. With an ingenious use of Sears, Roebuck catalogues as historical sources, and with the econometrics of hedonic price indices, Robert Gordon found that the rate of rise of the prices of producers’ goods (lathes, motors, and so forth) have like consumer goods been substantially overstated by not including their improving quality.23 In short, we’re much better off now compared to 1800 than the conventional measures of national product suggest.
- [back] For the international comparisons Maddison 2006, and in particular pp. 437, 443 for the factor of sixteen from 1700 to 2001 in international Geary-Khamis dollars of 1990. When a figure such as this is not footnoted it will regularly be lifted from Maddison's amazing oeuvre, such as 2007. For Britain itself see Feinstein 1972, Feinstein in Feinstein and Pollard 1988 and Crafts 1985.
- [back] Marglin 2007.
- [back] Clark 2007.
- [back] More 1516, p. 65.
- [back] All these figure from Maddison 2001 (in 2006), Appendix B, Table 21, p. 264.
- [back] Nordhaus 1997.
- [back] Boskin and others 1998. Gordon 2006 revises the figures up for one effect, down for another, leaving a 1.0 percent bias in the consumer price index. See the summary table on the earlier work in Moulton 1996.
- [back] Fogel 2008.
- [back] Maddison 2006, p. 265.
- [back] Friedman 2005.
- [back] Fogel 2002, pp. 1, 2, 4, 236.
- [back] Cox and Alm 1999, pp. 14-15.
- [back] Cox and Alm 1999, p. 43.
- [back] Fogel 2002, p. 266.
- [back] Compare Fouquet 2008.
- [back] Temple 1986 (2007), p. 89.
- [back] Maddison 2007, p. 320.
- [back] The point about Carnegie is made by Otteson 2006, p. 165.
- [back] Berg 1998, p. 140.
- [back] Sokoloff 1988; Sokoloff and Khan 1990.
- [back] Berg 1998, pp. 146-148.
- [back] Berg 1998, p. 154.
- [back] Gordon 1990.
