MS:
Let me add a bit to RK's discussion of deflation, Japan-style.
First, there are many measures of (macroeconomic) price changes; we can use not just personal consumption but also government purchases, business fixed investment, and imports and exports to get a broader measure. It moves much the same.
Second, and even wonkier, alongside the desire to preserve room to use monetary policy, measured rates of inflation overstate the level of price changes. So if we want actual inflation to be (say) 1%, then we need to aim higher. The main reason for this is that consumers (and for the broader measures, business and the government) substitute away from more expensive goods to less expensive substitutes.
To calculate inflation we need an initial basket of goods and the quantities purchased of each. We then can calculate how much income we need to purchase that basket. In each subsequent period, we can collect current prices and calculate whether we need more money
(inflation) or less (deflation) to buy the same basket.
If one good falls in price relative to others - which of course is happening all the time - then we move more of our budget in that direction. If a better new good comes along, we're better off, but the inflation measure doesn't capture that. Similarly, if a new good comes along that captures market share, presumably that happens because we're better off. But the inflation measure doesn't capture that. Finally, there's the actual process of collecting price data. As the retail sector shifts, and there's been a revolution in retailing in Japan since the mid-1990s. (Both cars and roadside malls became commonplace in rural and suburban
Japan.) Data collection inevitably lags such changes, but new retailers gain share because they offer value (which may include convenience and the selection of goods and simple [or, if you're trying to design a retail outlet, not so simple!] ambience. And online shopping is even harder to work with.
Note that one reason inflation does not matter is government debt: incomes and tax collections go up, too, not just interest payments. However, when there's deflation, the real rate of interest the government pays can't fall. This is another area of asymmetry, reflecting the "zero lower bound" of (nominal) interest rates.

mike smitka, economics
washington and lee university