The Consumer Price Index (specifically, the Consumer Price Index for All Urban Consumers) is the most common measure for inflation. But when attempting to predict how your own spending will change throughout the course of retirement, there are two reasons that historical CPI figures might not be a perfect predictor:
- Future changes in CPI may not be similar to past changes in CPI, and
- Your rate of spending change doesn’t necessarily match that tracked by the CPI.
What Does the CPI Track?
The CPI measures a specific bundle of goods, as determined by a survey of several-thousand families regarding their spending habits. Per the Bureau of Labor Statistics’ website, the following categories are included in the calculation. (Note: The items provided in each group are just examples. They’re not the only goods measured in that group. For the full list, see page 11 of this pdf.)
- Food and Beverages (breakfast cereal, milk, coffee, chicken, wine, full service meals, snacks)
- Housing (rent of primary residence, owners’ equivalent rent, fuel oil, bedroom furniture)
- Apparel (men’s shirts and sweaters, women’s dresses, jewelry)
- Transportation (new vehicles, airline fares, gasoline, motor vehicle insurance)
- Medical Care (prescription drugs and medical supplies, physicians’ services, eyeglasses and eye care, hospital services)
- Recreation (televisions, toys, pets and pet products, sports equipment, admissions)
- Education and Communication (college tuition, postage, telephone services, computer software and accessories)
- Other Goods and Services (tobacco and smoking products, haircuts and other personal services, funeral expenses).
Those may not be the goods that you spend money on. For example, as somebody who doesn’t own a car, doesn’t use any prescription drugs at the moment, doesn’t smoke, cuts his own hair, and has a rather unusual diet (vegetarian, gluten-free), the CPI isn’t exactly a perfect fit for my own spending habits.
Of course, you may look at my spending choices and think, “Wow, what a weirdo!” But that’s the point. We all spend money on different things.
Even if you do buy most of the goods/services tracked by the CPI, it’s entirely possible that your allocation between those goods/services is significantly different from the allocation used to calculate the index.
Consumption Changes vs. Price Changes
When attempting to determine how much income you need in retirement, it’s also important to remember that your consumption might change, not just the prices of the things you buy.
For example, the price of prescription drugs might increase by 5% per year, but if the rate at which you’re taking them is doubling every five years, then your own prescription drug spending will be growing faster than the rate of change that’s reported by the CPI.
Predicting Your Personal Rate of Inflation
My point here is not that the CPI is useless or that it’s way off base for everybody. Neither of those statements is true. But there is a chance that it won’t align very well with your own rate of spending change.
If you want the most precise answer possible as to how your spending will change over time, my suggestion would be to:
- Track your current spending (broken down by category),
- Do your best to estimate how your rate of consumption will change for each category, then
- Try to estimate how the prices for each category will change.
Hi. I'm Mike Piper, the author of this blog. I'm 
🙂 This was on my list of ideas to write an article about, Mike. An additional reason the CPI may not apply to you personally include your ability to live frugally and find things used rather than buying new all the time. I’m not sure of any resources you can use to estimate your own inflation rate – it was something I was going to research when I wrote the article. But it’s at least useful to realize that CPI may not fit anywhere close to your spending profile.
Hi Paul.
Good point regarding new vs. used buying habits.
As to estimating your own rate: To the extent that your own purchases are at least included in the CPI (though probably in different proportions), you can pull up the detailed price reports to see the historical change in prices for those items.
Of course, that still leaves you with the whole “past inflation rates =/= future inflation rates” problem.
Love this topic. When we were putting together a piece on inflation, came across a Reuters story on measuring inflation that made the interesting point that our perception of the inflation in the cost of items we purchase is often skewed. We all tend to focus more on increases in the cost of frequently purchased items (like in my case, coffee) and less on the slower rise or even decline in larger, but more rarely purchased items like technology. Sitting down and putting together a list of expenditures as you suggest seems a good way to help get around that natural inclination.
If interested here’s a link to the Reuters story: http://blogs.reuters.com/great-debate/2010/10/18/there-is-no-such-thing-as-inflation/
“We tend to focus more on increases in the cost of frequently purchased items and less on the slower rise or even decline in larger, but more rarely purchased items like technology.”
I’d never thought of that before, but it makes perfect sense to me. Thanks for sharing that article.
On estimating one’s personal rate of inflation you wrote:
“3. 3.Try to estimate how the prices for each category will change.”
Much more telling is to catagorize your personal spending data into the categries used by the Bureau of Labor Statistics to measure inflation. Then weight the actual inflation data according to how much you spend in each category.
The data is here: http://www.bls.gov/news.release/cpi.nr0.htm
Just.
Hi Justine. Good to see you here. 🙂
And you’re right–that does sound like a good method.