One of the strangest controversies in the economics profession is that there's some debate as to whether or not nominal shocks have real consequences. And it seems to me that a great deal of mischief is being done here by the use of the word "real" to mean "adjusted for inflation" when as we can see in the debate over whether Social Security benefits should be indexed to the CPI-W or the C-CPI-U there's no unambiguous definition of what the alleged real quantities are.
So stop and think for a minute about the economy. In the past month I have negotiated a new salary for 2013, signed a contract to buy a house, reached an agreement on a 30-year mortgage, signed a two-year mobile phone service contract with Verizon, and agreed to accept a fee for work I'll be doing in January. None of these arrangements is indexed to the CPI-W, the C-CPI-U, the CPI-U, the PCE deflator, the GDP deflator, or any other measure of inflation. Like the vast majority of economic agreements, they are specified in nominal terms. Adjusting for inflation to derive "real" values is an interesting economists' concept, applied to some public sector programs and union contracts but these different adjustments apply different price indexes and the majority of business is transacted nominally. So since business is conducted in nominal terms, nominal shocks impact the economic relationships of firms and households.
And one should note it's a good thing that the world is specified in nominal terms. If we inflation-adjusted everything, then relatively minor "real" shocks could have crazy consequences. A drought would push up the price of grain, and therefore food in general, and therefore there'd be inflation. Then thanks to the inflation everything would be indexed upward and there'd be . . . even more inflation.