• This is a political forum that is non-biased/non-partisan and treats every person's position on topics equally. This debate forum is not aligned to any political party. In today's politics, many ideas are split between and even within all the political parties. Often we find ourselves agreeing on one platform but some topics break our mold. We are here to discuss them in a civil political debate. If this is your first visit to our political forums, be sure to check out the RULES. Registering for debate politics is necessary before posting. Register today to participate - it's free!

Why CPI does not capture the inflation you feel

Perhaps you should explain why Owner's Equivalent of Rent is used instead of house prices.

You tell me, because it seems to be doing a terrible job at capturing the actual inflation of real estate.

The real reason the CPI doesn't represent any individual's experience is that it is the weighted average of thousands of goods. Nobody buys everything in the CPI every month and not in those proportions. And it is the average of Urban consumers...The entire state of Montana is excluded from the survey, as is most of New Mexico and at least one of the Dakotas.

I understand that there will be significant differences from place to place, but how is it that we're putting up with this large of an error?
 
You tell me, because it seems to be doing a terrible job at capturing the actual inflation of real estate.



I understand that there will be significant differences from place to place, but how is it that we're putting up with this large of an error?

With entitlement payments (e.g. social security and CSRS & military pensions) being a large portion of "mandatory" federal spending it is wise to have (make?) the "official" inflation level as low as possible. It costs many billions for each 1% of annual COLA raise that must be meted out by the federal government which, of course, then gets compounded by future COLAs over time.
 
Last edited:
You tell me, because it seems to be doing a terrible job at capturing the actual inflation of real estate.
The purpose of the Consumer Price Index is to measure the change in consumption goods. Investment items are excluded...so art, antiques, jewelry, stocks, bonds, etc are not included. Houses are a problem in that they are both a consumer good and an investment item. So how do we separate the investment portion and measure it as consumption? Instead of looking at the actual price, we look at the opportunity cost. If the owner were to rent the house instead of living in it, what would he be making from the rent? That is his consumption cost..the opportunity cost for not renting it. Of course, all the other parts of home ownership: utilities, repairs, appliances, furniture, etc are included in the CPI as well.


I understand that there will be significant differences from place to place, but how is it that we're putting up with this large of an error?
What error?
 
The inflation rate often used quality adjustments. Also item substitution when a particular item rises to much.

Examples

Your computer. An average computer in 1990 would have cost $2000 for a computer with 1/100 of the power of a $500 dollar computer. So for computing power the inflation rate is more than likely extremely negative. The same would be applied for any products with computers in them, from phones to card at less deflationary levels.
Since the object is to measure the price changes of the same items, if an item changes significantly, then it is no longer the same item and quality adjustment is applied. This works both ways: if a can of tuna goes from a 5oz can to a 4.8oz can, but the price doesn't change, BLS would record that as a price increase because you are paying more for less. I'm not sure why anyone would think quality adjustment is not a good thing.

The item substitution would see the T bone steak replaced with a blade steak then to ground beef then to pork as prices for specific items go up. As to when they would go back to a T Bone and how they would make the price adjustment I do not know.
That is absolute nonsense. What actually happens is that the Consumer Expenditure Survey determines how much of each item are purchased and each is assigned a weight based on Price X Quantity of each sub-category as a percent of Price X Quantity for each larger category. Then BLS Economic Assistants go out to the stores with an item list for each store, and the individual items are randomly selected proportional to sales. So T-bone, blade, filet mignon, NY strip, ribeye, etc are all sampled in a particular city and the price changes from the previous month are recorded, the weights applied, and we have the price change for steak in Seattle. Ground beef is a separate item category, so all the types of ground beef are sampled and aggregated. Then ground beef, steak, and roasts are aggregated for the beef category. Then pork, poultry, lamb (with all their sub-types aggregated) are all aggregated for the meat category. And then that's aggregated into Food at Home. The individual substitution you're talking about just doesn't happen.

Now, if a particular item is not in stock at the store, then the EA can choose a closely related item, in the same category, as a substitute. That entry is flagged, and an economist at BLS Headquarters in DC determines whether or not it can be a straight substitute, needs quality adjustment, or is a break in the series. Bottom round for top round would probably be a straight substitution, but ribeye for chuck would definitely not be. And you certainly cannot substitute pork for beef, except in the case of hotdogs.

The weights are adjusted periodically, to account for changing spending patterns, and item categories such as steak or ground beef or apples use the geometric mean to lower the effect of large price changes on the assumption that people will switch to an item that has not changed as much. The effect is actually pretty small.

The government has a strong interest to manipulate inflation rates both for social security and for social stability.

Sure, but the economists at BLS don't have any interest in manipulating inflation, and any changes to methodology take years to go into effect.

Full disclosure: I was formerly an economist at BLS, and I did have a short stint working on the CPI, determining item substitutions and quality adjustments for a couple of item categories.
 
The purpose of the Consumer Price Index is to measure the change in consumption goods. Investment items are excluded...so art, antiques, jewelry, stocks, bonds, etc are not included. Houses are a problem in that they are both a consumer good and an investment item. So how do we separate the investment portion and measure it as consumption? Instead of looking at the actual price, we look at the opportunity cost. If the owner were to rent the house instead of living in it, what would he be making from the rent? That is his consumption cost..the opportunity cost for not renting it. Of course, all the other parts of home ownership: utilities, repairs, appliances, furniture, etc are included in the CPI as well.

What error?

Do you have some evidence that growth in home prices has far exceeded rents?
 
Do you have some evidence that growth in home prices has far exceeded rents?

Since I haven't made that claim, why are you asking me? I haven't looked, so I have no idea. But again, if you're looking at home prices, you're looking at an investment portion.
 
Since I haven't made that claim, why are you asking me? I haven't looked, so I have no idea. But again, if you're looking at home prices, you're looking at an investment portion.

If rents are keeping up with home prices, then why aren't owners' equivalent rents?
 
If rents are keeping up with home prices, then why aren't owners' equivalent rents?

Well, you were using Case-Shiller home price index as your measure, so let's compare that with CPI rent of primary residence and owner's equivalent rent of primary residence:

fredgraph.png


The home price index simply does not measure the same thing as the CPI measures and we're looking at different samples, time periods, concepts, definitions, methodologies. You cannot compare them.

What I'm still wondering is why the thread title is about the CPI but you're only talking about shelter.
 
What I'm still wondering is why the thread title is about the CPI but you're only talking about shelter.

Because shelter is the largest individual component of it, and it's what people are spending the most money on.

As to your other points, we're looking at the CPI measure of rents. I don't think that it's keeping up with what median rents actually are. Other groups have recorded median rents and have found that they are keeping up with median home prices.
 
For years now people have been complaining about the CPI, that the inflation that they are experiencing is not adequately captured by the measure. I want to show in a few simple graphs why that is the case.
I shouldn't, but I will anyway ;)

You have this backwards. It's not that people are experiencing inflation that is not reflected in CPI. There really is almost no inflation right now. It's that people's experiences of economic realities are deeply flawed, and they perceive inflation that does not actually exist.

At least four well-known cognitive biases, and at least two economic conditions of recent years, plays into this:

1) Money Illusion. In a nutshell: People confuse nominal value of currency with purchasing power. E.g. when I was a kid, pizza cost $0.75 per slice; now, it's $2.50 or more. We feel that as inflation, but we ignore that our wages have risen during that time.

2) Bias Towards Negative: Human beings place more emotional weight on negative changes than we do on positive ones. When the price of gas falls 10¢, most people don't notice it for long, and they quickly adapt to the lower price. But if the price of gas goes up 10¢, people will notice -- and complain about it, and sometimes adjust their behavior, far more than the price cut.

3) Relative Status: People don't base much of their evaluations of their economic status based on absolutes. E.g. middle-class Americans are stupendously wealthy compared to most people in the world -- two cars, nice home, clean food, tons of entertainment, cheap clothing and so on. We mentally tally our condition compared to others, in particular those who are doing well, and this makes us discontent with our current status -- even if, in absolute terms, we're doing fairly well.

In more conventional terms, this is "Keeping Up With the Joneses." If there are 20 homes in your neighborhood, chances are good that 2 or 3 (or 10 or 15) of those homes are doing better than you at any given moment, and some are borrowing to buy fancy stuff that makes them feel fulfilled. This, combined with the Hedonic Treadmill (we adapt quickly to improvements in our absolute material/economic condition), can make you feel like you're behind the 8-ball not based on how well you're doing, but how you're doing compared to your perception of the neighbors.

4) Sorry for not remembering the technical name, but: People pay more attention to smaller, routine, negative stimuli than bigger, infrequent ones. Again, people will notice changes in things like gas prices or food prices -- even though that spending may only make up 15% of their budget -- because they're making those purchases several times a week. They are unlikely to properly calculate something like a major change to federal taxes, because it happens once a year, they aren't closely tracking it, may not remember last year's costs, and aren't calculating the effect on their regular paychecks (and/or they acclimate to them quickly).

5) Economic condition: Wages have been flat for almost the entire US population since the mid-70s. Even though that absolute level is miles beyond most of the world, we don't see that, and just see the Jones' new SUV.


Another key issue here: Housing costs have gone up, in part because the size of homes has increased dramatically. In fact, that explains pretty much all of the increase in housing costs.

housing1.png


housing2.png

New US homes today are 1,000 square feet larger than in 1973 and living space per person has nearly doubled - AEI

Equally important is that credit for mortgages is still near record lows. High housing prices don't hurt much when your rate is 4% and the cost is spread over 30 years.

Ultimately, homeowners aren't put out much by higher housing costs. People love it when their home values are on the rise. In fact, most of people's assets at retirement are their home.

Renters may feel a bit more pressure, but what typically happens is they are merely pushed out to lower-rent / less-desirable / less convenient areas. This has resulted in a significant increase in poor people living in suburbs, as "inner cities" become gentrified and desirable. (https://www.nbcnews.com/news/us-news/suburbs-house-more-poor-americans-inner-cities-n665386)

Sorry, but despite correctly noting that housing prices are a big part of CPI and have gone up, your analysis ultimately doesn't add up. Better luck next time.
 
I shouldn't, but I will anyway ;)

You have this backwards. It's not that people are experiencing inflation that is not reflected in CPI. There really is almost no inflation right now. It's that people's experiences of economic realities are deeply flawed, and they perceive inflation that does not actually exist.

At least four well-known cognitive biases, and at least two economic conditions of recent years, plays into this:

1) Money Illusion. In a nutshell: People confuse nominal value of currency with purchasing power. E.g. when I was a kid, pizza cost $0.75 per slice; now, it's $2.50 or more. We feel that as inflation, but we ignore that our wages have risen during that time.

We feel that as inflation because it is inflation. What you showed is that inflation is not necessarily a loss in income.

3) Relative Status: People don't base much of their evaluations of their economic status based on absolutes. E.g. middle-class Americans are stupendously wealthy compared to most people in the world -- two cars, nice home, clean food, tons of entertainment, cheap clothing and so on. We mentally tally our condition compared to others, in particular those who are doing well, and this makes us discontent with our current status -- even if, in absolute terms, we're doing fairly well.

For Millennials, that's largely no longer attainable.

4) Sorry for not remembering the technical name, but: People pay more attention to smaller, routine, negative stimuli than bigger, infrequent ones. Again, people will notice changes in things like gas prices or food prices -- even though that spending may only make up 15% of their budget -- because they're making those purchases several times a week. They are unlikely to properly calculate something like a major change to federal taxes, because it happens once a year, they aren't closely tracking it, may not remember last year's costs, and aren't calculating the effect on their regular paychecks (and/or they acclimate to them quickly).

Are you saying that people are not accurately capturing how much rents have risen recently?

5) Economic condition: Wages have been flat for almost the entire US population since the mid-70s. Even though that absolute level is miles beyond most of the world, we don't see that, and just see the Jones' new SUV.

Household wages have been flat despite many households adding a second income earner. Given that CPI is not capturing inflation well, especially over long periods of time, people feel poorer, and they are poorer. A simple ratio of home prices to incomes reveals that.

Another key issue here: Housing costs have gone up, in part because the size of homes has increased dramatically. In fact, that explains pretty much all of the increase in housing costs.

Rather, I'd argue that we're moving the average toward larger homes because only the richer among us can afford to buy homes. That said, it's really not that large of an increase in size, and it certainly doesn't explain all of the significant rise in home prices vs. incomes.

Ultimately, homeowners aren't put out much by higher housing costs. People love it when their home values are on the rise. In fact, most of people's assets at retirement are their home.

Renters may feel a bit more pressure, but what typically happens is they are merely pushed out to lower-rent / less-desirable / less convenient areas. This has resulted in a significant increase in poor people living in suburbs, as "inner cities" become gentrified and desirable. (https://www.nbcnews.com/news/us-news/suburbs-house-more-poor-americans-inner-cities-n665386)

Sorry, but despite correctly noting that housing prices are a big part of CPI and have gone up, your analysis ultimately doesn't add up. Better luck next time.

So home prices have risen, but it doesn't matter because homeowners are better off and who cares about renters. Solid.
 
As to your other points, we're looking at the CPI measure of rents. I don't think that it's keeping up with what median rents actually are. Other groups have recorded median rents and have found that they are keeping up with median home prices.

well, right there is one issue: the CPI doesn't measure median rents, it measures the mean change in rents. So you're not comparing the same thing.
 
We feel that as inflation because it is inflation. What you showed is that inflation is not necessarily a loss in income.
I think you missed the point. Perhaps I didn't explain it well.

When inflation is matched by increases in wages, nominal prices go up, but purchasing power remains the same. The problem is that people confuse the nominal prices with purchasing power. The illusion is that you think your purchasing power has declined, but it hasn't.

And of course, individual experience is mere anecdote. It's not a valid measure of anything except that person's feelings.

I'd also add that Americans haven't experienced high inflation since the 70s.

Thus, if someone under the age of 40 tells me that inflation is out of control, while CPI figures have been at record lows for 10 years in a row? I'm going with CPI.


For Millennials, that's largely no longer attainable.
Huh?

Millennials do have certain disadvantages compared to previous generations, mostly because a) they mostly started entering the workforce during the worst economic downturn since the 1930s (and that will affect lifetime earnings), and b) they are facing a more competitive work environment, which requires more education, which incurs more debt (but also increases lifetime earnings).

However, it's not like everyone aged 22 to 37 is homeless and have no prospects for economic survival. 85% have their own residence; they tend to live in areas where they don't need cars and/or can rely on car sharing or other car services; more young people have college degrees than ever before. I.e. let's not go overboard when reviewing the economic status of younger people.


Are you saying that people are not accurately capturing how much rents have risen recently?
I'm saying that people do not perceive their actual costs of living accurately.

Housing costs are routine, but changes are infrequent (once a year, or less). People are more likely to notice or feel any negative price fluctuations on purchases that they make several times a week, even if those purchases make up a relatively small part of their total spending.


Household wages have been flat despite many households adding a second income earner. Given that CPI is not capturing inflation well, especially over long periods of time, people feel poorer, and they are poorer. A simple ratio of home prices to incomes reveals that.
First, I see no valid argument that "CPI doesn't accurately reflect inflation." For example, you note that housing is about 40% of current CPI measures -- and that figure matches a lot of current research.

Second, and again: Housing prices match closely to the size of the homes, meaning housing is more expensive because we are living in larger homes.

Third, the number of workers in households hasn't actually changed all that much since the 70s. LFPR went from 60% in 1970, peaked at 67% in 2001, and has hovered around 62% since around 2014. We should note this means that, for most part, median household income did not fall at the exact same rate as LFPR. So more households with couples now have dual earners, but it's not like twice as many people are working in 2018 than in 1978.


I'd argue that we're moving the average toward larger homes because only the richer among us can afford to buy homes. That said, it's really not that large of an increase in size, and it certainly doesn't explain all of the significant rise in home prices vs. incomes.
Sorry, that doesn't work out either.

Homes have nearly doubled in size since 1970. It's gone from 551 sq ft/person to 1058 sq ft/person today. The 2nd chart points out how cost of housing is correlated to the average square foot per person.

Home ownership rates also haven't changed all that dramatically. For decades, it was in the 60-63% range; in the 90s, it started to run up, and peaked with the real estate bubble in 2004 at 70%. Since then, it's fallen and stabilized around 63%.


So home prices have risen, but it doesn't matter because homeowners are better off and who cares about renters. Solid.
Let's try this again.

Case-Shiller doesn't actually describe housing costs. It's an index of home sales. It doesn't tell us anything at all about rents. It doesn't directly measure interest rates, or property taxes, or the size of the property, or inventory.

You haven't shown that Owners' Equivalent Rent of Residence is wrong. All you've shown is that it doesn't correlate to Case-Shiller, which is not surprising, because C-S doesn't tell us the cost of housing, nor was it ever intended to do so.
 
I think you missed the point. Perhaps I didn't explain it well.

When inflation is matched by increases in wages, nominal prices go up, but purchasing power remains the same. The problem is that people confuse the nominal prices with purchasing power. The illusion is that you think your purchasing power has declined, but it hasn't.

And of course, individual experience is mere anecdote. It's not a valid measure of anything except that person's feelings.

I'd also add that Americans haven't experienced high inflation since the 70s.

Thus, if someone under the age of 40 tells me that inflation is out of control, while CPI figures have been at record lows for 10 years in a row? I'm going with CPI.

I understand the difference between inflation and purchasing power. There has clearly been inflation in home values. I don't think that anyone will deny that, though you may try to make the case that quality has increased. What I'm arguing isn't about inflation per se, but more about a decrease in relative purchasing power. Median incomes today cannot purchase homes like they used to in the past. This is the problem, and CPI isn't capturing it because home values are increasing faster than CPI reported values.

Huh?

Millennials do have certain disadvantages compared to previous generations, mostly because a) they mostly started entering the workforce during the worst economic downturn since the 1930s (and that will affect lifetime earnings), and b) they are facing a more competitive work environment, which requires more education, which incurs more debt (but also increases lifetime earnings).

However, it's not like everyone aged 22 to 37 is homeless and have no prospects for economic survival. 85% have their own residence; they tend to live in areas where they don't need cars and/or can rely on car sharing or other car services; more young people have college degrees than ever before. I.e. let's not go overboard when reviewing the economic status of younger people.

They also have way more debt than any previous generation and have the lowest homeownership rate of any recent generation. I don't know how you can ignore those two massive factors.

I'm saying that people do not perceive their actual costs of living accurately.

Housing costs are routine, but changes are infrequent (once a year, or less). People are more likely to notice or feel any negative price fluctuations on purchases that they make several times a week, even if those purchases make up a relatively small part of their total spending.

What argument do you use to say that they're wrong? Home prices are up. Median rents compared to median incomes are up. It's only CPI that isn't showing the increase. Why do you refuse to admit that the CPI measures might be wrong?

If you want to find me information that says that rents aren't way more expensive for most people today, then please provide it.

2015-11-18-1.PNG
 
First, I see no valid argument that "CPI doesn't accurately reflect inflation." For example, you note that housing is about 40% of current CPI measures -- and that figure matches a lot of current research.


Second, and again: Housing prices match closely to the size of the homes, meaning housing is more expensive because we are living in larger homes.


I've shown that rents have not been keeping up with incomes. Real median income has been stagnant. So either housing is getting more expensive, or income measures are wrong. Which is it?


Third, the number of workers in households hasn't actually changed all that much since the 70s. LFPR went from 60% in 1970, peaked at 67% in 2001, and has hovered around 62% since around 2014. We should note this means that, for most part, median household income did not fall at the exact same rate as LFPR. So more households with couples now have dual earners, but it's not like twice as many people are working in 2018 than in 1978.


The proportion of working age women active is 25% higher today than it was in 1978. It's not double, but it is substantial.


Sorry, that doesn't work out either.


Homes have nearly doubled in size since 1970. It's gone from 551 sq ft/person to 1058 sq ft/person today. The 2nd chart points out how cost of housing is correlated to the average square foot per person.


Home ownership rates also haven't changed all that dramatically. For decades, it was in the 60-63% range; in the 90s, it started to run up, and peaked with the real estate bubble in 2004 at 70%. Since then, it's fallen and stabilized around 63%.


Of course it works out. We have a significantly older population today than we did in 1970. Correspondingly, we ought to have a much higher home ownership rate, but we don't anymore. Home ownership being the same percentage as 1970, when median age was under 30, as today, when it is closer to 40, is actually pretty terrible.


Let's try this again.


Case-Shiller doesn't actually describe housing costs. It's an index of home sales. It doesn't tell us anything at all about rents. It doesn't directly measure interest rates, or property taxes, or the size of the property, or inventory.


You haven't shown that Owners' Equivalent Rent of Residence is wrong. All you've shown is that it doesn't correlate to Case-Shiller, which is not surprising, because C-S doesn't tell us the cost of housing, nor was it ever intended to do so.


It tells us the price you have to pay, before interest, to buy a house. That's going to be highly correlated to the monthly cost of a home. While lowering interest rates can help lower the cost, it doesn't get you over the down payment hurdle, which is pretty big for a lot of people today. What we have today is this weird situation where many people are priced out of the market despite record low interest rates. Why is that? Because the cost is simply too much.


Are you really fine with homeownership being the same today as it was in 1970, even with the median age being about 10 years higher? Is this good for society?
 
For years now people have been complaining about the CPI, that the inflation that they are experiencing is not adequately captured by the measure. I want to show in a few simple graphs why that is the case.

fredgraph.png


Here I have plotted the Case-Shiller National Home Price Index along with CPI, where both are indexed to be equal in 1990. What we see is that the growth in home prices has wildly outpaced CPI. In fact, home prices are about 33% higher than where CPI would expect them to be.

But CPI is weighted to represent how much people spend on certain items. Thus, around 40% of CPI is weighted toward housing. So why is CPI not capturing this? Because instead of just using home values, they use something called Owners' Equivalent Rent of Residence:

fredgraph.png


Look at that graph and you tell me how you think this measure has performed.

I thought that this would be interesting to talk about, and problems like this cause many Americans to ignore economic measures because they do not reflect the reality that they face.

Finally, I believe that this graph represents a major source of frustration for many Americans:

fredgraph.png


I have shown here a simple new home sales price to median family income indexed to 1975. By this metric, home prices are 50% higher today than they were in 1975 (and I am not even correcting for the much higher percentage of two income households of today). If you are wondering why many have become dissatisfied with our economy and are turning to more radical politics, I would assume that this is a major source.

THat's really interesting, I'm curious though. If inflation is a rise in the aggragate price level, what is the cause and perhaps more importantly is there an offset? Is there something that helps to balance out the increase in prices or should we just be looking on how our costs rise?
 
THat's really interesting, I'm curious though. If inflation is a rise in the aggragate price level, what is the cause and perhaps more importantly is there an offset? Is there something that helps to balance out the increase in prices or should we just be looking on how our costs rise?

Incomes ought to be keeping up with prices, but for the past 20 years they have not been. Instead, most income gains have been going to the wealthiest, which is why Kushinator can show that average wealth is keeping up with prices. Median incomes, however, are not, and that is because we are increasing inequality. Most Americans are not benefiting from our current economy. In fact, based on the charts I've showed, they're losing ground.
 
Back
Top Bottom