The best measure of inflation theoretically is changes in the money supply. If the money supply doubled overnight due to counterfeiting, for example, you would expect prices to double. However, the data on money supply is rather imprecise - it's not so simple to define money or track it accurately. Still, money supply growth is almost always way above the CPI. Because of productivity and economic growth, consumer goods should naturally fall in price over time. The difference between that fall and the actual nominal price increases we see is the rate of inflation, and will normally approximate to money supply growth over the long term. Gold and real estate follow their own cycles and are therefore poor measures of inflation over the short and medium0term. Gold fell significantly from 1980 to 2000 yet inflation was positive over that period. Real estate is falling now yet inflation is positive. Gold rose during the Great Depression yet inflation was negative. Over the long-term they are better indicators, since the supply of land and gold are fairly steady, the change in price should roughly correlate with the change in demand, which is heavily affected by the chance in the money supply. A good short & medium-term measure of inflation IMO is the change in nominal wages across the whole economy. Wages collapsed during the Great Depression, and soared during the 1970s stagflation. That was a strong correlation with inflation during those periods.