Sources and series: see piketty.pse.ens.fr/capital21c.
By adding these two curves, we can obtain a third curve showing the rate of growth of total global output (Figure 2.5). Until 1950, this had always been less than 2 percent per year, before leaping to 4 percent in the period 1950–1990, an exceptionally high level that reflected both the highest demographic growth rate in history and the highest growth rate in output per head. The rate of growth of global output then began to fall, dropping below 3.5 percent in the period 1990–2012, despite extremely high growth rates in emerging countries, most notably China. According to my median scenario, this rate will continue through 2030 before dropping to 3 percent in 2030–2050 and then to roughly 1.5 percent during the second half of the twenty-first century.
I have already conceded that these “median” forecasts are highly hypothetical. The key point is that regardless of the exact dates and growth rates (details that are obviously important), the two bell curves of global growth are in large part already determined. The median forecast shown on Figures 2.2–5 is optimistic in two respects: first, because it assumes that productivity growth in the wealthy countries will continue at a rate of more than 1 percent per year (which assumes significant technological progress, especially in the area of clean energy), and second, perhaps more important, because it assumes that emerging economies will continue to converge with the rich economies, without major political or military impediments, until the process is complete, around 2050, which is very rapid. It is easy to imagine less optimistic scenarios, in which case the bell curve of global growth could fall faster to levels lower than those indicated on these graphs.
The Question of Inflation
The foregoing overview of growth since the Industrial Revolution would be woefully incomplete if I did not discuss the question of inflation. Some would say that inflation is a purely monetary phenomenon with which we do not need to concern ourselves. In fact, all the growth rates I have discussed thus far are so-called real growth rates, which are obtained by subtracting the rate of inflation (derived from the consumer price index) from the so-called nominal growth rate (measured in terms of consumer prices).
In reality, inflation plays a key role in this investigation. As noted, the use of a price index based on “averages” poses a problem, because growth always bring forth new goods and services and leads to enormous shifts in relative prices, which are difficult to summarize in a single index. As a result, the concepts of inflation and growth are not always very well defined. The decomposition of nominal growth (the only kind that can be observed with the naked eye, as it were) into a real component and an inflation component is in part arbitrary and has been the source of numerous controversies.
For example, if the nominal growth rate is 3 percent per year and prices increase by 2 percent, then we say that the real growth rate is 1 percent. But if we revise the inflation estimate downward because, for example, we believe that the real price of smartphones and tablets has decreased much more than we thought previously (given the considerable increase in their quality and performance, which statisticians try to measure carefully—no mean feat), so that we now think that prices rose by only 1.5 percent, then we conclude that the real growth rate is 1.5 percent. In fact, when differences are this small, it is difficult to be certain about the correct figure, and each estimate captures part of the truth: growth was no doubt closer to 1.5 percent for aficionados of smartphones and tablets and closer to 1 percent for others.
Relative price movements can play an even more decisive role in Ricardo’s theory based on the principle of scarcity: if certain prices, such as those for land, buildings, or gasoline, rise to very high levels for a prolonged period of time, this can permanently alter the distribution of wealth in favor of those who happen to be the initial owners of those scarce resources.
In addition to the question of relative prices, I will show that inflation per se—that is, a generalized increase of all prices—can also play a fundamental role in the dynamics of the wealth distribution. Indeed, it was essentially inflation that allowed the wealthy countries to get rid of the public debt they owed at the end of World War II. Inflation also led to various redistributions among social groups over the course of the twentieth century, often in a chaotic, uncontrolled manner. Conversely, the wealth-based society that flourished in the eighteenth and nineteenth centuries was inextricably linked to the very stable monetary conditions that persisted over this very long period.
The Great Monetary Stability of the Eighteenth and Nineteenth Centuries
To back up a bit: the first crucial fact to bear in mind is that inflation is largely a twentieth-century phenomenon. Before that, up to World War I, inflation was zero or close to it. Prices sometimes rose or fell sharply for a period of several years or even decades, but these price movements generally balanced out in the end. This was the case in all countries for which we possess long-run price series.
More precisely, if we look at average price increases over the periods 1700–1820 and 1820–1913, we find that inflation was insignificant in France, Britain, the United States, and Germany: at most 0.2–0.3 percent per year. We even find periods of slightly negative price movements: for example, Britain and the United States in the nineteenth century (−0.2 percent per year if we average the two cases between 1820 and 1913).
To be sure, there were a few exceptions to the general rule of monetary stability, but each of them was short-lived, and the return to normal came quickly, as though it were inevitable. One particularly emblematic case was that of the French Revolution. Late in 1789, the revolutionary government issued its famous assignats, which became a true circulating currency and medium of exchange by 1790 or 1791. It was one of the first historical examples of paper money. This gave rise to high inflation (measured in assignats) until 1794 or 1795. The important point, however, is that the return to metal coinage, after creation of the franc germinal, took place at the same parity as the currency of the Ancien Régime. The law of 18 germinal, Year III (April 7, 1795), did away with the old livre tournois (which reminded people too much of the monarchy) and replaced it with the franc, which became the country’s new official monetary unit. It had the same metal content as its predecessor. A 1-franc coin was supposed to contain exactly 4.5 grams of fine silver (as the livre tournois had done since 1726). This was confirmed by the law of 1796 and again by the law of 1803, which permanently established bimetallism in France (based on gold and silver).27
Ultimately, prices measured in francs in the period 1800–1810 were roughly the same as prices expressed in livres tournois in the period 1770–1780, so that the change of monetary unit during the Revolution did not alter the purchasing power of money in any way. The novelists of the early nineteenth century, starting with Balzac, moved constantly from one unit to another when characterizing income and wealth: for contemporary readers, the franc germinal (or “franc-or”) and livre tournois were one and the same. For Père Goriot, “a thousand two hundred livres” of rent was perfectly equivalent to “twelve hundred francs,” and no further specification was needed.
The gold value of the franc set in 1803 was not officially changed until June 25, 1928, when a new monetary law was adopted. In fact, the Banque de France had been relieved of the obligation to exchange its notes for gold or silver since August 1914, so that the “franc-or” had already become a “paper franc” and remained such until the monetary stabilization of 1926–1928. Nevertheless, the same parity with metal remained in effect from 1726 to 1914—a not insignificant period of time.
We find the same degree of monetary stability in the British pound sterling. Despite slight adjustments, the conversion rate between French and British currencies remained quite stable for two centuries: the pound sterling continued to be worth 20–25 livres tournois or francs germinal from the eighteenth century until 1914.28 For British novelists of the time, the pound sterling and its strange offspring, such as shillings and guineas, seemed as solid as marble, j
ust as the livre tournois and franc-or did to French novelists.29 Each of these units seemed to measure quantities that did not vary with time, thus laying down markers that bestowed an aura of eternity on monetary magnitudes and a kind of permanence to social distinctions.
The same was true in other countries: the only major changes concerned the definition of new units of currency or the creation of new currencies, such as the US dollar in 1775 and the gold mark in 1873. But once the parities with metal were set, nothing changed: in the nineteenth and early twentieth centuries, everyone knew that a pound sterling was worth about 5 dollars, 20 marks, and 25 francs. The value of money had not changed for decades, and no one saw any reason to think it would be different in the future.
The Meaning of Money in Literary Classics
In eighteenth- and nineteenth-century novels, money was everywhere, not only as an abstract force but above all as a palpable, concrete magnitude. Writers frequently described the income and wealth of their characters in francs or pounds, not to overwhelm us with numbers but because these quantities established a character’s social status in the mind of the reader. Everyone knew what standard of living these numbers represented.
These monetary markers were stable, moreover, because growth was relatively slow, so that the amounts in question changed only very gradually, over many decades. In the eighteenth century, per capita income grew very slowly. In Great Britain, the average income was on the order of 30 pounds a year in the early 1800s, when Jane Austen wrote her novels.30 The same average income could have been observed in 1720 or 1770. Hence these were very stable reference points, with which Austen had grown up. She knew that to live comfortably and elegantly, secure proper transportation and clothing, eat well, and find amusement and a necessary minimum of domestic servants, one needed—by her lights—at least twenty to thirty times that much. The characters in her novels consider themselves free from need only if they dispose of incomes of 500 to 1,000 pounds a year.
I will have a lot more to say about the structure of inequality and standards of living that underlies these realities and perceptions, and in particular about the distribution of wealth and income that flowed from them. At this stage, the important point is that absent inflation and in view of very low growth, these sums reflect very concrete and stable realities. Indeed, a half century later, in the 1850s, the average income was barely 40–50 pounds a year. Readers probably found the amounts mentioned by Jane Austen somewhat too small to live comfortably but were not totally confused by them. By the turn of the twentieth century, the average income in Great Britain had risen to 80–90 pounds a year. The increase was noticeable, but annual incomes of 1,000 pounds or more—the kind that Austen talked about—still marked a significant divide.
We find the same stability of monetary references in the French novel. In France, the average income was roughly 400–500 francs per year in the period 1810–1820, in which Balzac set Père Goriot. Expressed in livres tournois, the average income was just slightly lower in the Ancien Régime. Balzac, like Austen, described a world in which it took twenty to thirty times that much to live decently: with an income of less than 10–20,000 francs, a Balzacian hero would feel that he lived in misery. Again, these orders of magnitude would change only very gradually over the course of the nineteenth century and into the Belle Époque: they would long seem familiar to readers.31 These amounts allowed the writer to economically set the scene, hint at a way of life, evoke rivalries, and, in a word, describe a civilization.
One could easily multiply examples by drawing on American, German, and Italian novels, as well as on the literature of all the other countries that experienced this long period of monetary stability. Until World War I, money had meaning, and novelists did not fail to exploit it, explore it, and turn it into a literary subject.
The Loss of Monetary Bearings in the Twentieth Century
This world collapsed for good with World War I. To pay for this war of extraordinary violence and intensity, to pay for soldiers and for the ever more costly and sophisticated weapons they used, governments went deeply into debt. As early as August 1914, the principal belligerents ended the convertibility of their currency into gold. After the war, all countries resorted to one degree or another to the printing press to deal with their enormous public debts. Attempts to reintroduce the gold standard in the 1920s did not survive the crisis of the 1930s: Britain abandoned the gold standard in 1931, the United States in 1933, France in 1936. The post–World War II gold standard would prove to be barely more robust: established in 1946, it ended in 1971 when the dollar ceased to be convertible into gold.
Between 1913 and 1950, inflation in France exceeded 13 percent per year (so that prices rose by a factor of 100), and inflation in Germany was 17 percent per year (so that prices rose by a factor of more than 300). In Britain and the United States, which suffered less damage and less political destabilization from the two wars, the rate of inflation was significantly lower: barely 3 percent per year in the period 1913–1950. Yet this still means that prices were multiplied by three, following two centuries in which prices had barely moved at all.
In all countries the shocks of the period 1914–1945 disrupted the monetary certitudes of the prewar world, not least because the inflationary process unleashed by war has never really ended.
We see this very clearly in Figure 2.6, which shows the evolution of inflation by subperiod for four countries in the period 1700–2012. Note that inflation ranged between 2 and 6 percent per year on average from 1950 to 1970, before rising sharply in the 1970s to the point where average inflation reached 10 percent in Britain and 8 percent in France in the period 1970–1990, despite the beginnings of significant disinflation nearly everywhere after 1980. If we compare this behavior of inflation with that of the previous decades, it is tempting to think that the period 1990–2012, with average inflation of around 2 percent in the four countries (a little less in Germany and France, a little more in Britain and the United States), signified a return to the zero inflation of the pre–World War I years.
To make this inference, however, one would have to forget that inflation of 2 percent per year is quite different from zero inflation. If we add annual inflation of 2 percent to real growth of 1–2 percent, then all of our key amounts—output, income, wages—must be increasing 3–4 percent a year, so that after ten or twenty years, the sums we are dealing with will bear no relation to present quantities. Who remembers the prevailing wages of the late 1980s or early 1990s? Furthermore, it is perfectly possible that this inflation of 2 percent per year will rise somewhat in the coming years, in view of the changes in monetary policy that have taken place since 2007–2008, especially in Britain and the United States. The monetary regime today differs significantly from the monetary regime in force a century ago. It is also interesting to note that Germany and France, the two countries that resorted most to inflation in the twentieth century, and more specifically between 1913 and 1950, today seem to be the most hesitant when it comes to using inflationary policy. What is more, they built a monetary zone, the Eurozone, that is based almost entirely on the principle of combating inflation.
FIGURE 2.6. Inflation since the Industrial Revolution
Inflation in the rich countries was zero in the eighteenth and nineteenth centuries, high in the twentieth century, and roughly 2 percent a year since 1990.
Sources and series: see piketty.pse.ens.fr/capital21c.
I will have more to say later about the role played by inflation in the dynamics of wealth distribution, and in particular about the accumulation and distribution of fortunes, in various periods of time.
At this stage, I merely want to stress the fact that the loss of stable monetary reference points in the twentieth century marks a significant rupture with previous centuries, not only in the realms of economics and politics but also in regard to social, cultural, and literary matters. It is surely no accident that money—at least in the form of specific amounts—virtually disappeared fro
m literature after the shocks of 1914–1945. Specific references to wealth and income were omnipresent in the literature of all countries before 1914; these references gradually dropped out of sight between 1914 and 1945 and never truly reemerged. This is true not only of European and American novels but also of the literature of other continents. The novels of Naguib Mahfouz, or at any rate those that unfold in Cairo between the two world wars, before prices were distorted by inflation, lavish attention on income and wealth as a way of situating characters and explaining their anxieties. We are not far from the world of Balzac and Austen. Obviously, the social structures are very different, but it is still possible to orient perceptions, expectations, and hierarchies in relation to monetary references. The novels of Orhan Pamuk, set in Istanbul in the 1970s, that is, in a period during which inflation had long since rendered the meaning of money ambiguous, omit mention of any specific sums. In Snow, Pamuk even has his hero, a novelist like himself, say that there is nothing more tiresome for a novelist than to speak about money or discuss last year’s prices and incomes. The world has clearly changed a great deal since the nineteenth century.
PART TWO
THE DYNAMICS OF THE CAPITAL/INCOME RATIO
{THREE}
The Metamorphoses of Capital
In Part One, I introduced the basic concepts of income and capital and reviewed the main stages of income and output growth since the Industrial Revolution.
In this part, I am going to concentrate on the evolution of the capital stock, looking at both its overall size, as measured by the capital/income ratio, and its breakdown into different types of assets, whose nature has changed radically since the eighteenth century. I will consider various forms of wealth (land, buildings, machinery, firms, stocks, bonds, patents, livestock, gold, natural resources, etc.) and examine their development over time, starting with Great Britain and France, the countries about which we possess the most information over the long run. But first I want to take a brief detour through literature, which in the cases of Britain and France offers a very good introduction to the subject of wealth.
Capital in the Twenty-First Century Page 13