How Do We Measure Inflation?

Vernier calipers measuring metal nut on schematic


Inflation has been in the news a lot this year. There is continual discussion of the potential impact of Federal Reserve actions and federal government spending on inflation.

Consumers worry about inflation and its impact on their purchasing power. Conceptually, it’s easy to understand inflation. However, our personal experiences of inflation may not align with either the concept or with inflation as it is reported in the news.

Inflation is rising. What does that mean? Does anyone know?

The concept of inflation is closely related to the assessment of living standards and the “cost-of-living.” Economists (and everyone else) want to know if living standards are improving. This should be easy to measure – just count the goods and services that people consume each year. If the counts are higher this year than last, people are better off.

To make this concrete, imagine someone who subsists entirely on chicken and rice, 200 pounds of chicken and 2,000 pounds of rice, for example. If they buy 20% more chicken and rice this year than last, we might say that their living standard has increased by 20%.

Cost of living comparison

Rick Miller, Sensible Financial

The table provides details, and the graph illustrates. Both chicken and rice quantities are 20% larger, as is the budget. That translates into a 20% living standard improvement, with no inflation, as prices haven’t changed. [The solid orange line represents this year’s budget at this year’s prices (and passes through this year’s basket), while the dotted blue line represents last year’s budget at last year’s prices (passing through last year’s basket).]

Living standard increase year over year

Rick Miller, Sensible Financial

Measuring living standards and inflation isn’t usually this easy, unfortunately. Economists who try to figure out if people are better off now than they were before and by how much must address three issues. First, people don’t buy goods and services in fixed proportions. For example, suppose our someone buys 50% more chicken this year than last and 10% less rice. Are they better or worse off? By how much? We could answer, “somewhere between 10% worse off and 50% better off,” but that isn’t precise or satisfying.

Spending differences based on different quantities

Rick Miller, Sensible Financial

In our example, spending would have increased by 20%. With no change in prices, we might argue that their living standard has increased 20%. By the way, this is the basic concept behind the measurement of gross domestic product (GDP) – we measure economic progress as the change in the value of goods and services produced each year.

Graph overview

RICk MILLER, Sensible Financial

Now the second problem – price increases. Suppose all prices rise by the same percentage (say 10%). This is inflation, an increase in the general level of prices, a rise in the cost-of-living (a decline in the purchasing power of the dollar). Since both prices rose by 10%, it’s intuitive that inflation is 10%.

Comparison of quantities and spending year over year

Rick Miller, Sensible Financial

We estimate inflation by calculating how much the cost-of-living has increased, that is, comparing how much we’d have to spend this year vs last year to buy last year’s basket of food (200 pounds of chicken and 1,000 pounds of rice). We can also estimate inflation by calculating how much it would have cost this year vs last year to buy this year’s basket of food (300 pounds of chicken and 900 pounds of rice). Both inflation estimates are 10%.

Since total spending increased by 32%, and the cost of living increased by 10% (inflation), living standard would have increased by 20% (32% -10% – 2% “large change adjustment” – see box below). We already knew this from our analysis of spending change with no change in prices.

Formula to determine living standard percentage change

Rick Miller, Sensible Financial

Finally, the third problem: prices for goods and services don’t always change in the same proportion. In our final example, we’ll see what happens when chicken prices rise by 10% while rice prices rise by 50%. The quantities are the same as before.

The table shows that spending rises by 50%. We can calculate the change in the cost of living in two ways. We can ask how much more it costs to buy last year’s basket at this year’s prices – $1,300 vs $1,000 or 30%. Or we can ask how much more it would cost to buy this year’s basket paying this year’s prices rather than last year’s – $1,500 vs $1,200 or 25%.

Spending and Inflation, large price increases

Rick Miller, Sensible Financial

We can see this on the graphs that follow, also. We calculate living standard changes (solid arrows) keeping prices the same and changing the baskets (moving from a dashed line to a solid line of the same color). We calculate inflation (dotted arrows) keeping the baskets the same and changing prices (moving from blue to orange for dashed or solid lines).

There are two paths to move from last year’s basket and last year’s prices to this year’s basket and this year’s prices: 

  • Start with prior year prices and calculate the improvement in living standard: how much more we can buy this year ($1,000 for last year’s basket to $1,200 for this year’s basket – a 20% improvement). Then calculate inflation as the difference in cost of this year’s basket moving from prior year prices to current year prices ($1,200 to $1,500 – a 25% increase).

Inflation based on current basket

Rick Miller, Sensible Financial

  • Start with the prior year’s basket and calculate inflation as the difference in cost from the prior year’s prices to this year’s prices ($1,000 to $1,300 – 30%). Then calculate the living standard increase as the cost increase from the prior year’s basket to this year’s basket with the current year’s prices ($1,300 to $1,500 or 15%).

Inflation based on last year\’s basket

Rick Miller, Sensible Financial

Which inflation calculation should we use?

Which inflation estimate is “right?” Is inflation 25% or 30%? In general, using this year’s quantities tends to produce a lower inflation estimate, as people tend to increase their spending on items with smaller price increases (also called relative price decreases). That’s chicken in this example – its 10% price increase is smaller than either inflation estimate. People tend to reduce spending on items with higher (or relative) price increases (rice’s 50% price increase is larger than either inflation estimate).

Unfortunately, there is no single correct answer. An economist named Laspeyres invented the measure using last year’s basket, and another named Paasche invented the other. Both are well accepted estimation methods. In fact, there are many ways to estimate inflation. As just one example, we could average the Laspeyres and Paasche estimates to construct a third estimate.

Inflation is relative.

Our simple example illustrates the difficulty of disentangling relative price changes from inflation. If we spoke with someone who subsisted solely on chicken, they would say that inflation is overstated – the price they care about most has increased less (only 10%) than the reported inflation rate of 25% or 30%. On the other hand, someone who buys only rice would say that the reported inflation rate is much too low. The most important price to them has risen about twice as fast (50%)!

Thus, when we hear something like “health care inflation is rising faster than reported inflation,” we can interpret that as “health care prices are rising faster than other prices.” That could happen even if inflation were zero and the purchasing power of the dollar wasn’t changing as a whole.

Estimating inflation is significantly more complex than this simple example suggests. Besides choosing the base period (prior or current year), we must decide what basket or budget we will use to estimate inflation. In my next article, I’ll delve into the CPI (Consumer Price Index) and the PCEPI (Personal Consumption Expenditure Price Index).

The foregoing content reflects Rick Miller’s opinions and is subject to change at any time without notice. Content provided herein is for informational purposes only and should not be used or construed as investment advice or a recommendation regarding the purchase or sale of any security. There is no guarantee that the statements, opinions, or forecasts provided herein will prove to be correct.  

Past performance may not be indicative of future results. Indices are not available for direct investment. Any investor who attempts to mimic the performance of an index would incur fees and expenses which would reduce returns.  

Securities investing involves risk, including the potential for loss of principal. There is no assurance that any investment plan or strategy will be successful.