The simplest evidence that CPI substantially understates inflation is the decrease in standard of living and purchasing power of seniors who have lived for a few years with their total income only based on Social Security payments.
In simpler terms, anyone who believes that the CPI is accurate needs to look at how much the actual standard of living of those living only on Social Security has dropped over the last 5-10 years, when using real apples to apples comparisons.
The late 2012 proposal to use "Chained CPI" instead of CPI-W for Social Security inflation adjustments is yet another proof for the doubters that the Boskin Commissions adjustments in 1998-9 had the same purposes - understate the real inflation rate to save money, at the expense of stable purchasing power for seniors on Social Security.
BLS medical cost increases have been about 2% lower then total medical cost since 2000.
That 2% is about 50% of the difference between the BLS figures and the total actual increases per the CMS and Kaiser.
Hedonics, substitution bias, geometric weighting, cost of living
The BLS CPI is not a cost of living index, where a certain standard of living is maintained and measured, but rather measures price changes via a market basket approach. It was originally designed to be much closer to measuring a stable & consistent standard of living, but has drifted towards a market basket based index starting around 1968 with Medicare and continuing with the housing OER adjustments in 1982 and the Boskin Commission recommendations in 1996-1999, etc.
In other and more direct words, it no longer measures real inflation via BLS best efforts, but rather measures what people spend as they change their buying habits when they can no longer afford their previous standard of living for whatever reasons. In spite of that, the BLS continues to assert that the CPI measures inflation ("The CPI is the most widely used measure of inflation" and "The CPI and its components are used to adjust other economic series for price changes and to translate these series into inflation-free dollars." Source).
The Boskin Commission, which completed its work in 1996. focused on three main areas; hedonics or quality improvements, substitution and geometric weighting.
Quality improvements or hedonics:
Yes, TVs are better today than 10 years ago with internet connectivity, HDMI, etc... but no one can buy a new one without those improvements, which means its partially an actual and unaccounted for price increase. And then we have the wild variable and probable over-valuing of the various improvements. The repair record vs. older analog sets also does not seem to be as good, but we're unable to find any reliable data on either side.
Another example, the computer price CPI index component was lowered by about 6.5% in 1998, but did work get done 6.5% faster? We don't know since we have not found any data or information on how the adjustment is calculated. Should the fun or entertainment component of a new computer really be a part of a price index that should measure changes in standards of living, and be as much as 6.5%?
In January 1999, the BLS began treating price increases due to legislatively mandated modifications to goods and services for pollution control purposes as price increases. Before this change, such price increases were regarded as a change in quality and were excluded from the CPI. We believe this one to be wise, but what about all the missing price increases prior to 1999?
Another example about cars - since the late 60s, actual car prices have increased about 600%, but the BLS hedonics adjustments drop the price increase to about 300%. The problem seldom mentioned is that the actual dollar price one pays is 600%, not 300%.
Reverse hedonics: while we can certainly agree that the improvement from outhouses to indoor plumbing and similar changes are a quality improvement and deserve some weight in a true inflation index, we also note that the opposite should also be done. In other words, reverse adjustments should also be done for things like smaller airline seats with less leg room (source) (Canada too), ( You Won’t Believe What Airplane Food Looked Like In The ‘60s ), plastic veneer and particle board dining room tables instead of solid wood, and the many other items which have had substantial quality losses over the decades. And yes, cars are better now... but also cost much more to repair when something goes wrong and also cost much more percentage wise to insure. There is no quality adjustment process for mercury in your fish, pesticides on your fruits and vegetables, or antibiotics and hormones in your meat and poultry. Is your neighborhood noisier or smellier than it used to be?
Geometric weighting and substitution bias:
The geometric estimator takes into account that from one period to the next consumers may change the quantity that they purchase of a particular item within a category like meat because of a change in the relative price of the item. In other words, if beef prices go up and chicken doesn't, the BLS assumes that more chicken is consumed, and that is regardless of whether the item now being consumed represents a decrease in the standard of living.
The geometric adjustments are done on about 60% of the entire set of CPI components, but exclude items like rent, electricity, cable TV, doctors' services, eyeglasses and eye care.
In January 1999, BLS implemented a new housing sample of rental units and an estimation method for homeowners' housing costs, which uses only data on rental units. Before this change, the housing sample included owner-occupied and rental units, and homeowners' housing costs were estimated from the costs experienced by renters whose units were comparable to owner-occupied units in the housing sample.
While a case can be made for any of the Boskin adjustments, especially hedonics (including reverse hedonics), substitution and geometric weighting should not apply if one is trying to measure true changes in a stable standard of living or a cost of living index, aka a real and complete measure of inflation.
Also note that the BLS estimates that the CPI is affected by the Boskin related changes by about 1.1% as of 1999. We believe that it is substantially higher now (2012).
Products remaining at the same price, but are offered in smaller quantity or lower quality, or both.
The lack of reverse hedonics factors have been noted elsewhere, and they should include what we give up in the way of spare time, privacy, spare time and ethics/morality issues.
A 2"x4" piece of lumber keeps getting smaller over the years and is now 1.5"x3.5"
College fees (Worcester State University in Massachusetts "parking/pedestrian fee", At the University of California Santa Cruz, where tuition runs to nearly $35,000 for non-residents, students every year pay more than 30 additional fees 2014 including a small charge for what's billed as "free" HIV testing. Students at Oklahoma State University pay 18 different fees, including a "life safety and security fee." A $100 "globalization fee" exists at Howard University. At the University of Massachusetts Amherst, the flagship of the UMass system, mandatory fees are more than six times the cost of in-state tuition. Planned obsolescence, examples being cars that are only designed to last 100,000 miles or vacuum cleaners that no longer usually last 10 years or more
No taxes are included anywhere in the calculation of the CPI in spite of the BLS statements "The CPI is the most widely used measure of inflation" and
"The CPI and its components are used to adjust other economic series for price changes and to translate these series into inflation-free dollars." Source. Taxes are prices that consumers actually pay, and again CPI stands for Consumer Price Index.
Also note that our CPPI does not include any adjustments for taxes (or additional and substantial deficit spending), which of course makes it lower than it should be per its attempt to measure real inflation.
The total tax amounts have been going down the last few years (as of 2012) too, which can account for many thinking that inflation is lower than it actually is.
How can a price index that states it measures actual inflation not count total taxes, which have been as high as 33% of income since 1913?
Could any of our conclusions be incorrect? Of course, but until such time as the BLS fully opens all their databases and makes public all the adjustments and algorithms etc., there is no way to be certain (let alone all the issues with creating a reasonably accurate price index). We don't expect the databases etc. to be opened or made public, if for no other reason than "national security".
Lastly, the list of items above are not complete and there are more, but rather than confuse the overall issue even more we've elected to stop here. The major items that are missing besides taxes are asset related (stocks, bonds, commodities, etc.).