To adapt the exchange between Ernest Hemingway and his literary agent: the super-rich are very different from the rich. They have even more money.
Much of the traditional analysis of inequality focuses on the relationship between those towards the middle and those towards the top of the income range.
But in recent years the most dramatic widening, at least in the US, has been between the very top and everyone else.
The share of income claimed by the top 1 percent of American earners declined after the Great Depression and the second world war, but from the late 1970s it suddenly started to rise. Meanwhile, wage growth for many in the middle has stalled.
Between 1976 and 2007 in the US, 58 percent of the total growth in income was captured by the top 1 percent.
Not only that, but the trend appeared to be accelerating; that figure was 45 percent during the 1990s economic expansion under President Bill Clinton, but rose to 65 percent from 2002-2007 under President George W. Bush. And the super-super-rich within the super-rich have been doing better still.
Between 1990 and 2005, a full 2 percentage points of the 2.3 percent increase in the overall share of the top 1 percent of earners went to the top 0.1 percent.
Working out why is one of those debates in which academic economists fumble through murky data to construct tentative conclusions, which are then seized as weapons in an intense political debate about fairness and equality.
The argument has raged for years about whether the super-rich have got that way because they are super-talented, super-lucky or just super-manipulative.
The super-talented explanation is that a few extraordinarily talented people have simultaneously emerged in fields – investment banking, corporate management, law – and are reaping rewards commensurate with their productivity.
A more sophisticated argument is that, while the dispersal of talent is not necessarily different, globalisation and other structural changes have pushed many industries toward a “winner-take-all” pattern, where small differences in talent at the top can make huge differences in reward.
During the financial boom of the late 2000s, huge pay packages for chief executive officers such as Dick Fuld at the now defunct Lehman Brothers and Chuck Prince, formerly of Citigroup, were justified on the grounds that the investment banking world had become ever more competitive and productive.
This second explanation generates an exciting tale about a new club of international superstars – the top dogs on a dog-eat-dog planet, the Davos elite.
A clear example of the latter is in sports. But it may not be applicable elsewhere. For one, the pattern of inequality at the top has not followed some immutable global law of economics.
The very rich became richer in most English-speaking countries, even those with more redistributive tax systems and a stronger commitment to equality as an end in itself than in the US.
But for continental European countries and Japan, the share of the top 1 percent flat-lined over the same period.
Sceptics of economic efficiency suggest a third, more cynical explanation: that the very rich, mainly CEOs and top finance executives, have found ways to pay themselves undue amounts, often by tying compensation to the stock market during a boom.
They would point to the fates of Lehman and Citigroup as evidence that large salaries were not justified. The main constraint on this is public, stockholder and employee tolerance of such behaviour.
Robert Lawrence, a Harvard professor and expert on the effect of globalisation on inequality, says: “The debate is still unresolved between the competing explanations: does inequality represent greater economic efficiency or a weaker ‘outrage restraint’ on executive pay?” Some circumstantial evidence supports the “super-manipulative” theory.
Prof Lawrence says that US corporations have not become particularly more globalised over the past quarter century.
Changing technologies may have facilitated the rise of winner-take-all industries, even within domestic economics.
But the globalisation argument looks more like an ex-post justification than an actual explanation.
The key question now is what happens as a result of the financial crisis. If the super-rich consolidate their rising share of income, it will run counter to previous trends.
It will also provide more ammunition to those arguing that the rising share of the very rich is more to do with the ability to distort and manipulate markets.
Early indications are that, with profits high and much of high earners’ pay still based on the stock market, inequality at the very top shows few signs of falling off.
Prof Lawrence says: “There has been an unusually high profit share [in gross domestic product]?...?I presume the claimants on those profits are going to be doing pretty well.” Given what brought the economy to this pass, such an outcome is unlikely to be greeted with quite the same general sense of resignation that met the rise of the super-rich in America in the past quarter century.