Will We See Deflation in the Next 12 Months?

December 12, 2021

I’m not worried at this point about a deflationary spiral, but I see what, to my view, is a plausible scenario where the CPI actually goes negative in the next twelve months. I go through the categories and my predictions component by component below, but there are four main items driving the story that I’ll mention here.

First, I assume a sharp reversal in new and used car prices. The 11.1 percent increase in the former and 31.4 percent increase in the latter, have added 1.5 percentage points to the inflation rate over the last year. This run-up is due to the well-known shortage of semiconductors. It seems that manufacturers are overcoming this shortage and getting up to normal production levels. This may lead to a situation where they are not only meeting normal demand, but actually could be overproducing and needing to markdown prices.

A second big assumption is a sharp moderation in food prices. The price of store-bought food has risen by 6.4 percent over the last year, adding 0.5 percentage points to the inflation rate (food bought at restaurants added another 0.4 percentage points). This has been driven by a huge surge in demand, where we seem to be eating more of everything. We also see supply chain problems raising shipping costs.

I am betting on the surge in demand easing somewhat and the supply chain problems being resolved over the course of the year. In the past, sharp run-ups in food prices have been followed by declines or periods of very slow growth. I’m betting on the latter.  

My third assumption is a sharp reduction in gas and other energy prices, reversing some of the recent run-ups. Gas prices increased 58.1 percent in the last year, adding 2.4 percentage points to the inflation rate.

I assume a partial reversal of this run-up, with a drop in gas prices simply reflecting the recent drop in world oil prices. That would imply an 18 percent decline in prices from the November level, knocking 0.8 percentage points off of the inflation rate for the next twelve months.

Finally, I assume that the prices of many other items, where we have seen a sharp run-up due to supply chain issues, such as appliances and furniture, will level off in the next year as these problems get resolved.

My model here is televisions. The index for televisions had been falling for decades, but it surged by 10.2 percent from March to August (a 26.3 percent annual rate). Since August, the index for televisions has fallen sharply in the last three months, dropping by more than 4.0 percent. I expect that we will see a similar story with many other items in the year ahead. This reversal may come soon if it turns out that many stores over-ordered for the holiday shopping season.

Before going into the item-by-item assessment, I’ll add a point that is worth repeating. The bond markets seem to agree with the view that the inflation we have been seeing is temporary. The interest rate on a 10-year Treasury bond on Friday was under 1.5 percent. That put the breakeven inflation rate for an inflation-indexed bond and the conventional 10-year bond at less than 2.5 percent. (If we allow for the 0.2-0.4 percentage point difference between CPI inflation and inflation as measured by the personal consumption expenditure deflator, this is pretty much in line with the Fed’s 2.0 percent target.) Obviously, investors in the bond market are not expecting anything like 6.8 percent inflation to persist or even 4-5 percent inflation.

This should be somewhat reassuring, but as someone who was warning about both the stock bubble in the 1990s and the housing bubble in the 2000s, I know financial markets can be wrong. But it is still worth paying some attention to what people with money on the line are doing.

Inflation: November 2021 to November 2022

I can’t claim to have a crystal ball that tells me what inflation in the different components will be over the next year, but there is some basis for making reasonable guesses. So here is my story. I welcome corrections/additions by people who are more knowledgeable about specific areas.[1]

          Projected  Contribution
      Inflation   Inflation   of  
  (weights)   Nov 20-Nov 21 Nov 21-Nov 22 Component
All items 100   6.8   -0.54      
Core 78.536   4.9   0.43      
Food at home 7.733   6.4   1   0.08  
Food away from home 6.262   5.8   2   0.13  
Energy commodities 4.207   57.5   -18   -0.76  
Energy services 3.262   10.7   -10   -0.33  
New vehicles 3.856   11.1   -11   -0.42  
Used cars and trucks 3.35   31.4   -33.5   -1.12  
Motor vehicle parts and equipment 0.401   10.2   -1   0.00  
Motor vehicle maintenance and repair(1) 1.085   4.9   4   0.04  
Motor vehicle insurance 1.557   5.7   0.5   0.01  
Airline fares 0.596   -3.7   20.1   0.12  
Other Transportation services 1.774       0   0.00  
Alcoholic beverages 0.997   1.9   1.9   0.02  
Tobacco and smoking products 0.615   8.9   3.9   0.02  
Shelter 32.425   3.8   3.5   1.13  
Household furnishings and supplies 3.774   6   0   0.00  
Household operations 0.89   8.4   5.5   0.05  
Water and sewer and trash collection services 1.074   3.5   3.5   0.04  
Apparel 2.725   5   0   0.00  
Recreation commodities 1.961   3.9   -2.5   -0.05  
Recreation services 3.703   2.8   3.5   0.13  
Medical care commodities 1.493   0.2   0.5   0.01  
Medical care services 7.002   2.1   3   0.21  
Education and communication commodities 0.48   0.9   -3.6   -0.02  
Education and communication services 6.043   1.7   1.7   0.10  
Personal care products 0.643   -0.2   -0.2   0.00  
Miscellaneous personal goods 0.195   6   -1   0.00  
Other personal services 1.632   4.5   4.5   0.07  

Source: Bureau of Labor Statistics and author’s calculations.

Gasoline and Other Energy

Higher gas prices have featured front and center in the story of runaway inflation impoverishing the masses. The good news here is that we can be pretty certain that prices will decline. The price of oil fell to ridiculously low levels in the pandemic (futures prices were actually negative). They then soared to more than $83 a barrel at the start of November as the economy reopened. They have since fallen back to $71 a barrel.

The surge in oil prices led to a huge jump in gasoline prices, which were up 58.1 percent over the last year. I’m betting on an 18.0 percent decline over the next year. This is simply taking where the CPI gas index was back in October of 2018 when the price of oil was roughly at its current level.

I don’t know whether oil prices will go higher or lower from today forward, but there is a good reason to expect the general direction will be downward. We know Biden and the Democrats are doing horrible things to the fossil fuel industry (imposing environmental regulations and restricting where they can drill) but the reality is that demand for fossil fuels is likely to be falling over the next decade.

Electric car sales are growing rapidly here and around the world. Tesla alone projects that it will be selling more than 20 million cars a year by 2030, a number that is almost 20 percent larger than the current U.S. car market. Take that with the appropriate amount of salt, but it seems likely that in the not distant future, most of the cars being sold will be electric.

As this switch takes place, the demand and price of fossil fuels are likely to fall. When producers look out to this future, many are likely to make the bet that it is better to get something for their oil today than risk having it still in the ground twenty or thirty years from now when there may be very little demand. This logic is likely to be especially important for big OPEC producers, like Saudi Arabia, that have very low marginal costs for bringing oil to the market.

Anyhow, that prediction on oil prices is obviously very speculative, but I’ll just put down my -18.0 percent for gas based on the current price. I’m applying this figure to the larger category of energy commodities, since this is mostly gas and the other items have closely tracked gas prices.

For energy services, I’m putting in a projection of a 10.0 percent price decline, largely reversing a 13.3 percent increase since the start of the pandemic. Higher natural gas prices were clearly a big factor in this run-up, and natural gas prices have also been falling sharply in recent weeks. The pattern over the last dozen years has been that sharp increases in this category were quickly followed by sharp declines. The index for energy services was just about 5.0 percent higher before the pandemic than it was a decade earlier.


I can’t say I have a good idea where food prices are going, primarily because I don’t really know what has caused them to go up so much, 6.4 percent over the last year for the food at home category. There is the obvious point that we seem to be eating a lot more food, but the question is why. Purchases of food for home consumption were 11.1 percent higher (adjusted for inflation) in the third quarter of 2021 than in the fourth quarter of 2019. By contrast, food purchases were just 2.8 percent higher in the fourth quarter of 2019 than they had been seven quarters earlier.

Part of this story is that people were buying less food at restaurants, but by the third quarter we were almost back to our pre-pandemic levels of restaurant sales, and by now we are above them, although somewhat below trend. So, are we really eating that much more food than before the pandemic and will that pattern continue?

And, just to be clear, we see the increase in every category. Real purchases of cereal and bakery products are up 14.4 percent, meat consumption 6.4 percent, dairy products 11.7 percent, with consumption of fruits and vegetables rising by the same amount.

I have no idea as to whether people will keep buying so much more food, but it doesn’t seem like a healthy development. Anyhow, for the path of inflation, I’m just going to assume that it follows past patterns. Where we have seen sharp price increases, they have generally been reversed or at least followed by periods of slow price growth.

Food prices rose by 6.6 percent from December 2007 to December 2008. They then fell by 2.4 percent the following year. They rose 6.0 percent from December 2010 to 2011, then rose just 1.3 percent the following year. In the decade before the pandemic began, they rose an average of 1.3 percent annually. After their 6.1 percent rise last year, I’ll put down a 1.0 percent increase over the next twelve months.

Restaurant prices have generally risen by about a percentage point more than food prices, presumably reflecting rising higher labor costs. The opposite has been true over the last year, with restaurant prices going up 5.8 percent, but I’ll assume this pattern resumes over the next year. I’ll put down 2.0 percent for the projected increase in restaurant prices.

Cars and Trucks

New and used cars have been an enormous factor in the inflation we have seen over the last year. Used vehicle prices rose by 31.4 percent and contributed 1.1 percentage points to the inflation rate over the last year. New vehicle prices rose by 11.1 percent and added 0.4 percentage points to the inflation rate.

The reason for these extraordinary price increases is hardly a secret, a fire in a major semiconductor factory in Japan has led to a worldwide shortage of semiconductors. This has led to major reductions in auto production in factories around the world.

While there is still a shortage of semiconductors, several major manufacturers report being back up to capacity. It is reasonable to expect that most factories will be running near capacity within a few months and the auto market will be close to normal by November of next year.

New vehicle prices are up 12.6 percent since the pandemic began. In the seven years from February 2013 to February 2020, they increased by a total of just over 1.0 percent. I’m going to assume that in the next year prices will return to something like their former path. I’m putting down a price drop of 11.0 percent.

Used vehicle prices are up 33.5 percent since the pandemic began. The used vehicle index had actually been falling in the years prior to the pandemic. I will assume that the increase since the pandemic began is reversed over the next year.

Motor Vehicle Equipment and Parts

Prices in this component rose 10.4 percent in the last year after rising less than 1.0 percent annually over the decade prior to the pandemic. This reflects both supply chain issues and also the increased demand for parts as people sought to improve used cars for sale or their own use.

With car production returning to normal, and supply chain issues coming under control, I expect some of this rise to be reversed. I am putting down -1.0 percent for the next year.

Motor Vehicle Repair and Maintenance

Inflation in this component rose to 4.9 percent over the last year, up from 3.5 percent over the prior year. Some of this is undoubtedly due to supply chain disruptions associated with reopening, as well as higher labor costs. I’m assuming that inflation in this component will fall back to 4.0 percent in the next year.

Car Insurance

The index for car insurance had been rising rapidly early in the last decade, but slowed sharply in the years just before the pandemic. In the two years prior to the pandemic, it increased by an average of 0.5 percent. It then fell sharply in the pandemic only to then rise rapidly as the economy reopened, going up 5.7 percent over the last year.

It is important to recognize that the CPI uses a gross measure for auto insurance, counting premiums rather than administrative costs and profits, as it does with health insurance. The sharp rises earlier in the last decade were mostly due to higher payouts. If payments for damages and medical expenses are under control, then the rise in premiums is likely to be limited. I assume that the rise in the next year will be 0.5 percent.

Airline Fares

Airfares plummeted at the start of the pandemic. They have recovered to some extent, but they are still 20.1 percent below their pre-pandemic level. Assuming that the pandemic is under control, it is likely that fares will recover to their pre-pandemic level. I’m putting in a 20.1 percent increase in airfares over the next year.

Other Transportation Services

This is a hodgepodge that includes inner-city and intercity bus travel, state licensing fees, and car rentals. I’m putting down a prediction of no change based on the fact that I expect the 37.2 percent increase in car rental prices to be largely reversed in the next year. This component comprises almost exactly 10 percent of the whole category, so a sharp decline in rental car price should be sufficient to offset increases in the other components.

Alcohol and Tobacco

The index for alcoholic beverages rose 1.9 percent over the last year. This is pretty much in line with its average over the prior five years. I will put down 1.9 percent for next year.

Tobacco prices are driven largely by state and local taxes on tobacco. In the last year, they rose by 8.9 percent. This is considerably more rapid than the 3.9 percent average increase over the prior decade. I am assuming that the rate of increase slows to its prior average.

Rent and Shelter

The two rental components of the CPI, rent proper and owners’ equivalent rent (OER) for owner-occupied housing, are huge factors in determining inflation. Together they account for 31.1 percent of the overall index and 39.6 percent of the core CPI.

The rate of rental inflation slowed in 2020, but has accelerated as the economy reopened. The rent proper index has risen 3.0 percent over the last year, while the OER index has risen by 3.5 percent. 

We have seen an interesting pattern develop since the pandemic began. Rents in high-priced areas are showing lower growth, while low-priced areas are seeing rapid rises. In the New York City metropolitan area, rents rose by 0.1 percent over the last year. In San Francisco, they fell by 0.5 percent. In Boston, rents are up 1.1 percent, and in DC by 0.2 percent. By contrast, Detroit rents were up 5.8 percent, in Atlanta 7.5 percent, and St. Louis 4.8 percent.

My guess is that this divergence continues, as workers with increased opportunities to work from home move to lower-priced parts of the country. That’s likely good news for most of the country – more affordable housing in expensive cities and a boost to growth in cities that had been previously left behind – but it’s not clear how it affects overall rental inflation.

One positive is that the rise in house prices during the pandemic, coupled with extraordinarily low interest rates, has led to a boom in housing construction. We’re on a path to having almost 1.8 million housing starts in 2021, up from less than 1.4 million in 2019. This is a positive development, but in a country with over 140 million housing units, an additional 400,000 is not going to have much impact on rents.

I will assume that both indexes return to roughly their pre-pandemic rates of inflation. I’m putting in 3.5 percent as my projection of inflation. This category also includes hotels. That index is currently 8.9 percent above its pre-pandemic level. This component accounts for less than 3.0 percent of the shelter index. I don’t expect that it will diverge enough from the 3.5 percent figure I’m putting down for rent to substantially alter the shelter projection.

Household Furnishings and Supplies

This component had a big spike in inflation in the last year, rising by 6.0 percent last year after having an average increase of less than 0.5 percent over the five years prior to the pandemic. This is the supply chain story. Many of the items in this category are imported, and even domestically produced items are tied up in transit. (It includes appliances.) As supply chain problems ease, there should be some price reversal. I expect that we will see prices in this category be roughly flat in the next year. I’m putting down no change.

Household Operations

This is a category that includes items like gardening and domestic workers. The index for “domestic services” rose 10.2 percent over the last year. (This is probably one reason why we are hearing so much about inflation in the media.) Overall, the index for household operations rose by 8.4 percent over the last year, presumably reflecting higher pay for workers in this sector. 

It is likely that low-paid workers will continue to receive substantial pay increases in the year ahead. I’m putting down 5.5 percent for this category.

Water and Sewer and Trash Collection Services

Prices in this category rose by 3.5 percent last year. This likely reflects higher wages for many workers. We are likely to see increases of roughly the same size in the year ahead. I’m putting down 3.5 percent.


Apparel prices rose by 5.0 percent last year, after falling by 5.1 percent in the prior year. The general direction for apparel prices has been downward, with the February 2020 index about 4.0 percent lower than its level from five years earlier. I will assume the index stays flat, although the sharp rise in the dollar over the last year would be a factor that should lower apparel prices.

Recreation Commodities

This category includes many items caught up in the supply chain. My favorite example is televisions. As noted earlier, the index for televisions had been falling for decades, but then rose 10.2 percent from March to August (a 26.3 percent annual rate). They have fallen sharply the last three months, although are not yet back to their March level.

The index for this category as a whole had been falling consistently for the decade prior to the pandemic at more than a 2.0 percent annual rate. It rose 3.9 percent in the last year. I expect the prior trend to return. Prices should fall by roughly 2.5 percent in the next year.

Recreation Services

Inflation in this category has been very contained, in large part because the pandemic has hugely depressed demand. The index rose by just 2.7 percent over the last year, roughly the average increase over the prior five years. It is reasonable to expect some pick-up in this measure over the next year, both because demand will increase as pandemic fears ease, and because many of the low-paid workers in the sector will get higher wages. I’m putting down 3.5 percent.

Medical Care Commodities

This category is primarily prescription drugs. After rising rapidly earlier in the century, it has slowed sharply in recent years. The index rose by just 0.2 percent over the last year. This likely had more to do with political pressures than the pandemic.

Medical Services

Inflation in medical services has been very limited in the last year, with the index rising just 2.1 percent. This is a sharp slowing from its immediate pre-pandemic pace (it had risen 5.5 percent in the prior year), but from 2015 to 2020, it had risen by an average of just 3.4 percent. 

There are factors pushing inflation in medical service prices in both directions going forward. On the one hand, many people put off care during the pandemic and will likely be making appointments when they feel more comfortable going to a medical facility. On the other hand, to be somewhat morbid, many of the people who were most in need of services died during the pandemic.

There is also the spread of telemedicine, which was far more widely adopted as a result of the pandemic. This should help to put downward pressure on prices. It is also likely that there will be considerable political pressure from the Biden administration and Congress to contain costs.

I am going to assume that, on net, this leaves us with a somewhat lower rate of inflation in health care services than we saw the prior five pre-pandemic years. I’m putting down 3.0 percent.

It is important to realize that most of the growth in prescription drug prices is not captured in the CPI index, since it only measures the prices of drugs currently on the market. If a new drug comes out carrying a price of $55,000 for a year’s dosage (like Aduhelm, the new Alzheimer’s drug), it does not affect the CPI. However, if the price declines in years down the road, due to new competition or going off-patent, this drop will show up in the index.

I will assume that this index rises by 0.5 percent over the next year.

Education and Communication Commodities

This includes a variety of items such as textbooks, computers, and smartphones. The index rose 0.9 percent last year, after falling at an average annual rate of 3.6 percent over the prior decade. I am assuming that this rate of decline resumes in the next year. I’m putting down -3.6 percent.

Education and Communication Services

This component combines college tuition and child care with telephone and Internet service. The former categories have seen modest price increases in the years before the pandemic, while communication services have generally been declining in price. College tuition growth is likely to be restrained in the near future as many schools will still rely to a large extent on remote learning and feel a need to restrain tuition increases. Phone and Internet providers may also feel some need to restrain price increases in response to political pressure.

Inflation in this component was 1.7 percent over the last year. I have assumed that it will be 1.7 percent again in the year going forward.

Personal Care Products

This category includes items like shaving cream, toothpaste, and shampoo. The index fell by 0.2 percent last year, roughly in line with past patterns. I will put down a decline of 0.2 percent for the next year.

Other Personal Services

This category includes a wide range of items like haircuts, legal services, and tax preparation. It rose by 4.5 percent last year, up from an average close to 2.5 percent in prior years. This presumably reflects more rapid pay growth for many of the lower-paid workers in this category. I am putting down 4.5 percent for next year.

Miscellaneous Personal Goods

I don’t know what these are. The index rose by 6.0 percent last year. Prices had been falling by an average of a bit more than 1.0 percent annually in the decade before the pandemic. I will assume that the rise last year was due to supply chain problems and that the decline will resume next year. I’m putting down -1.0 percent for this category.

Are We Good on Inflation?

I tried to use a critical eye in putting down these numbers. Some, like gas prices, clearly have a more solid foundation than others. I’m sure someone could justify different and higher numbers in each category, but these are my best guesses with the information I have. I welcome comments and criticisms.








[1] Careful observers will note that my weights only add up to 99.735. (They are actually “relative importance,” but we’ll leave that for another day.) I’m obviously missing some component that has a weight of 0.265 in the index. Suggestions welcome.


Support Cepr


If you value CEPR's work, support us by making a financial contribution.

Si valora el trabajo de CEPR, apóyenos haciendo una contribución financiera.

Donate Apóyanos

Keep up with our latest news