How to measure inflation

FEW pieces of economic data matter more at the moment than inflation. Markets follow the statistics obsessively, looking for clues to how economies are doing and how central bankswill respond. Rich-world inflation figures are diverging. In May the euro-zone inflation ratesank to just 0.5%, leading to speculation that the European Central Bank might begin printing money to buy up government bonds, lest prices begin falling. In Britain, inflationtumbled to 1.5% in May, down from 1.8% in April, even as house prices soared to new heights. In America, by contrast, inflation jumped to 2.1% in May by one measure, leading to fears that the Federal Reserve was waiting too long to raise its interest rates (though by another measure it held lower, at just 1.6%). What exactly is inflation and how is it measured?

The notion of inflation is clear enough: it is the rate of change in the level of prices in an economy. More specifically, it is a change in the value of money. When there is more money about—more of it coursing through an economy as giddy consumers hit the shops—bulging supply pushes down the value of each unit of money (each dollar or pound) and more of them are needed to buy things. When consumers hold their money more tightly, however, each unit of money is more valuable and prices fall, a phenomenon called deflation. This relationship between the flow of money and changes in prices is why central banks focus so intently on inflation. Reliable inflation figures are critical for other reasons as well: to compare American incomes in 1970 with those in 2010, for example, one must know how the value of a dollar changed over that time. And contracts or pensions often need to be adjusted to take account of inflation, to ensure that the sums involved do not shrink over time in real terms.

Most inflation data are gathered from surveys. In America, the Bureau of Labour Statistics records the prices of tens of thousands of goods each month. Prices are aggregated into sub-indices such as food, energy, apparel and health care, then averaged together, with the categories weighted according to their share of spending. Some indices focus on the prices of firms' inputs rather than the cost of consumer goods. Statistical agencies usually produce “headline” inflation indices, which include all sub-sectors, as well as “core” indices that omit things like food and energy, whose prices are volatile. Central banks focus on core inflation, which they reckon is the better predictor of future trends. Not all data gathering is done by governments. Many banks and research organisations also track price changes, and experiments like the Billion Price Project glean data from online retailers, allowing for real-time inflation updates.

Though rates differ from index to index, central banks can have reasonable confidence in the accuracy of the statistics when similar trends show up across the various measures. Still, the data-gatherers often face criticism. When prices of some goods rise, consumers may switch to others—from beef to chicken, for instance, if beef prices soar. If statisticians maintain the same weightings when such shifts occur, they overstate the amount that household budgets are being squeezed by rising prices. Yet simply weighting chicken more heavily fails to capture the fact that it is (to many households at least) an inferior product to beef. Such questions of quality often trouble the number-crunchers. The prices of some electronic goods, such as computers, have changed little in decades, yet the products are vastly better. Attempts to make “hedonic” adjustments to account for this are tricky. But with ever more data being collected and published, economists are getting better at capturing movements in prices. Where inflation indices are concerned, at least, the public can count on getting its money’s worth.

lest: 以免，与in fear of近义

tumble: 暴跌

soar：飞涨

giddy: 眩晕的 

bulge：隆起

indice：指数

glean：慢慢地收集