Thursday, December 23, 2021

The Ghost of Christmas Inflation

This is part of an ongoing series of essays on inflation.  This one is at Project Syndicate. The next post is somewhat longer and more academic with the same themes. 

The Ghost of Christmas Inflation

Inflation continues to surge. From its inflection point in February 2021 to last month, the US consumer price index has grown 6% – an 8% annualized rate. 

The underlying cause is no mystery. Starting in March 2020, the US government created about $3 trillion of new bank reserves (an equivalent to cash) and sent checks to people and businesses. The Treasury then borrowed another $2 trillion or so and sent even more checks. The total stimulus comes to about 25% of GDP, and to around 30% of the original federal debt. While much of the money went to help people and businesses severely hurt by the pandemic, much of it was also sent regardless of need, intended as stimulus (or “accommodation”) to stoke demand. The goal was to induce people to spend, and that is what they are now doing. 

Milton Friedman once said that if you want inflation, you can just drop money from helicopters. That is basically what the US government has done. But this US inflation is ultimately fiscal, not monetary. People do not have an excess of money relative to bonds; rather, people have extra savings and extra apparent wealth to spend. Had the government borrowed the entire $5 trillion to write the same checks, we likely would have the same inflation. 

Other purported factors – including “supply shocks,” “bottlenecks,” “demand shifts,” and corporate “greed” – are not relevant to the overall price level. The ports would not be clogged if people were not trying to buy lots of goods. If people wanted more TVs and fewer restaurant meals, the price of TVs would go up and the price of restaurant meals would go down. Greed did not suddenly break out last year. 

By contrast, inflation, when all prices and wages rise together, comes from the balance of overall supply and demand. The economy’s capacity to produce goods and services turns out to be lower than expected. Here, the labor shortage – the “Great Resignation” – is a key underlying fact. Employers can’t find people to work because many people remain on the sidelines, not even looking for jobs. 

The US Federal Reserve was completely surprised by the surge of inflation, and through most of the year insisted it would be “transitory,” and go away on its own. That turned out to be a major institutional failure. Is it not the Fed’s main job to understand the economy’s supply capacity and fill – but not overfill – the cup of demand? 

One might expect that among the thousands of economists the Fed employs, there is a group working on figuring out ports’ capacity, the effects of microchip shortages, how many people have retired or are not returning to work, and so forth. One would be disappointed. Central banks have sketchy ideas of supply, mostly centered on statistical trends in labor markets. 

Why did this fiscal stimulus produce inflation when previous stimulus efforts from 2008 to 2020 fizzled? There are several obvious possibilities. First, this stimulus was much bigger. Former US Secretary of the Treasury Lawrence H. Summers correctly prophesied inflation in March [Update: February] 2021 by simply looking at the immense size of the spending packages, relative to any reasonable estimate of the GDP shortfall. 

Second, officials misunderstood the COVID recession. GDP and employment did not fall because there was a lack of “demand.” In a pandemic, you can send people all the money in the world and they still won’t go out to dinner or book a flight, especially if those services are suspended by government fiat. To the economy, a pandemic is like a blizzard. If you send people a lot of money when the snow is falling, you do not get activity in the snowdrifts, but you will get inflation once the snow has cleared. 

Third, unlike in previous crises, the government created money and sent checks directly to businesses and households, rather than borrowing, spending, and waiting for the effect to spread to incomes. 

Will inflation continue? Fundamentally, inflation breaks out when people do not think the government will repay all its debts by eventually running fiscal surpluses. People then try to get rid of the debt and buy things instead, which drives up prices and lowers the real value of debt to what people believe the government will repay. Given that prices have risen 6%, people evidently believe that of the 30% debt expansion, the government will not repay at least 6%. If people believe that less of the debt expansion will be repaid, then the price level will continue to rise, as much as 30%. But inflation will eventually stop: A one-time fiscal helicopter drop leads to a one-time rise in the price level. 

So, whether inflation will continue depends on future fiscal and monetary policy. Fiscal policy is the big question: Now that we have crossed the Rubicon of people believing that a fiscal expansion will not be fully repaid, will people think the same about additional persistent deficits? The danger here is obvious. 

If fiscal inflation does erupt, containing it will be difficult. If monetary policymakers try to curtail inflation by raising interest rates, they will run into fiscal headwinds as well as a political buzz saw. First, with the debt-to-GDP ratio above 100%, if the Fed raises interest rates five percentage points, interest costs on the debt will rise by $1 trillion – 5% of GDP. Those interest costs must be paid, or inflation will just get worse. Similarly, if the European Central Bank raises interest rates, it increases Italy’s debt costs, threatening a new crisis and imperiling the ECB’s vast portfolio of sovereign bonds. 

Second, once inflation works its way to higher bond yields, stemming inflation requires higher fiscal surpluses to repay bondholders in more valuable dollars. Otherwise, inflation does not fall. 

Monetary policy alone cannot contain a bout of fiscal inflation. Nor can temporary “austerity,” especially sharply higher marginal tax rates that undercut long-run growth and therefore long-run tax revenues. The only lasting solution is to get the governments’ fiscal house in order. 

Finally, supply-oriented policy is needed to meet demand without driving up prices, to reduce the need for social spending, and, indirectly, to boost tax revenues without a larger tax base. Given supply constraints from regulations, labor laws, and disincentives created by social programs, potential solutions here should be obvious. 


A somewhat longer piece on the same themes is on its way, with documentation of the numbers and more explanation. The points here are also covered in more detail in the "covid inflation" chapter of Fiscal Theory of the Price Level.  


  1. What is your explanation for low interest rates on government securities? If people are not convinced that the government will run surpluses in the future to repay debts, why are they holding the bonds?

    1. The Federal Reserve is the catalyst behind the ultra-low rates by monetizing the debt issued by treasuries. A good person to follow on this is @robinbrooksiif on twitter, the fed's holding of gov debt since GFC has skyrocketed while foreign ownership has plummeted. So it's not that "people" are willing to hold negative bearing gov bonds, it's the FED has stepped in at all costs to ensure the debt gets held, pushing down rates in the process.

    2. The bond market is predicting a recession

    3. Treasury yields price the Fed reaction function - nothing more.

  2. I don't think it is entirely relevant whether fiscal policy or monetary policy is the more powerful driver. We have received double-barreled stimulus on both fronts.

    In normal times it is the job of the central bank to do the fine tuning. If fiscal stimulus is excessive, the Fed can be restrictive. But, these are not normal times. But, it does feel like the central bankers are asleep.

    Given the number of times I have tried to buy something in the last 18 months and found that it is simply not available anywhere at any price, I am surprised we have not seen more inflation already.

  3. "If people believe that less of the debt expansion will be repaid, then the price level will continue to rise, as much as 30%."

    As much as 30% if you think there won't be other X-trillion-dollar packages... These deficits might be changing peoples' perception of future deficits too.

    I'm afraid what you are doing over there in the US will soon arrive around here

  4. "Had the government borrowed the entire $5 trillion to write the same checks, we likely would have the same inflation."

    As MIT '84, this troubles me. It's the new theory, I know. And it isn't willingness to hold money vs. bonds that's driving things. But is it maybe debt that will be repaid vs. debt that will not? The Chinese aren't buying debt now. Nor are Americans. Just the Fed, the buyer of last resort.

    1. This is a deep comment, which I've been thinking about for a while. The next essays will have more of this. If reserves are debt people do not think will be repaid and treasury debt is debt people think will be repaid, that makes a big difference.

    2. This comment has been removed by the author.

  5. Do you have any falsifiable inflation predictions? Are we in for a 30% price level increase? More? Less?

    Personally, I've been more convinced so far by the tilt toward goods from services explaining a great deal of the inflation we've seen. As such, I suppose I'd be on team transitory with a baseline expectation that inflation will normalize to the acceptable sub 4% range over the next year. What are your expectations?

  6. Fundamentally, inflation breaks out when people do not think the government will repay all its debts by eventually running fiscal surpluses.--John Cochrane

    I dunno. Suppose I sell hot dogs from stand, $1 each. Things are in balance, I sell out by the end of the day, and my cart holds only 100 hot dogs.

    Then, demand starts to creep up. I start to sell out before the end of the day.

    So, I take advantage and raise prices, and get back into balance.

    Vice-versa, if I can afford to accept a lower price.

    When do I ponder Federal Reserve policy or government deficits? Don't I set prices daily based on demand?

    Also, if the Fed buys up a large amount of Treasuries...then are people assured the federal government can meet debt payments? After all, the interest payments on the Fed-owned Treasuries flow back into the Treasury.


    1. But why did demand for your hot dogs go up? Maybe people want fewer hamburgers? OK, but then the price of hamburgers goes down. How can people possibly spend more money on everything, to drive up the price of everything? Only if people are trying to get rid of money, and government debt in general. Aggregate demand is not demand, inflation is not a relative price change.

    2. Both Hot Dogs and Hamburgers “originate” with the same livestock,, more or less.
      Reasonable to assume that the Hamburger Guys are low on supply too, and, of course raising prices…
      My fear, being realized more and more every day, is that ~10% inflation is already baked-in across the board.

    3. This comment has been removed by the author.

    4. The price of everything can be driven up if the supply of everything is driven down. Imagine an isolated economy where some natural disaster destroys 10% of all business facilities of every kind: supply would go down and prices would go up.

      In this case facilities weren't destroyed: but lockdowns, reduced production and widespread supply chain disruptions and the price of widely used commodities like oil rose and widely drove up the price of other things. Fiscal stimulus led some people to have more funds to spend to allow prices in general to rise rather than merely leading some prices to fall while others rose.

      The blog post said: "If people wanted more TVs and fewer restaurant meals, the price of TVs would go up and the price of restaurant meals would go down."

      Except its more complicated than that: since covid led to people staying home more and buying goods for home use (I hadn't checked on whether TVs were included) and fewer meals out. However: even if competition would normally tend then to drive down restaurant meal prices, other factors like cost to comply with covid restrictions, reduced occupancy allowed, and worker shortages or workers demanding essentially a hazard pay premium for such jobs worked against that and applied to all the industry participants.

      The pandemic decimated small businesses: oddly I hadn't seen good data on that since relief programs did temporarily keep some afloat and there are supposedly a rise in some new businesses but I hadn't checked data (many I suspect are just individuals who went freelance when unemployed, either to actually get income or to collect the new unemployment for freelancers).

      Decimating small businesses led to less competition for some niches and more ability temporarily for oligopolies to increase prices in some niches or resist decreases in others.

      Oddly people talk about the high profits in some public companies: but thats a misleading picture since smaller companies are a different story. It is true that some large companies benefited from the decimation of smaller ones, and of course some profited from the pandemic increasing demand for some things.

      However also: companies with hope for the future often reinvest revenue into expansion rather than taking a profit. Sometimes profit taking means they are "saving for a rainy day" rather than investing in expansion or fixing supply chain issues, etc. Small companies are likely doing the same.

      The production capacity needs to shift to deal with the shift in demand to restrain price increases: and that requires capital being spent, often borrowed capital.

      Yet last I checked (AIER had a post about it) bank lending hasn't recovered, or last I checked investment is still not up to where it should be even though the shift in demand and the decimated businesses should require capital to rebuild and fix the supply chain mess. Again by analogy: picture it like a natural disaster physically destroyed 10% of business facilities and they needed to be rebuilt: capital would be needed.

      Of course instead: the government wants to take resources and focus on its programs rather than letting the private sector rebuild itself. Risk averse investors from overseas will turn to treasuries and keep those interest rates low, while not enough capital flows in to where its needed to rebuild which is more risky with an unpredictable pandemic going on and the economic situation uncertain. Investors that do take risks tend to be partly distracted with the "everything bubbles", stock prices detached from fundamentals, crypto, etc.

  7. "Given supply constraints from regulations, labor laws, and disincentives created by social programs, potential solutions here should be obvious."

    I'm no Marxist, but I do believe in equitable compensation and treatment of labor.

    So, I'm curious: which ones do we get rid of?

    My old classes in Labor History point to a reality that when firms and labor play nice, there are productivity gains to be earned and shared, for it reduces churn, helps morale, and builds trust.

  8. Is it really a one time helicopter drop if the government is running primary deficits financed by the FED? Its not as if the US has been comitted to 0 debt growth this whole time and the share of the debt being paid by borrowed funds appears to be shrinking.

  9. Excellent analysis as usual. Would be helpful if the writer were to address the failure of TIPS and longs bonds to price this in.

  10. This comment has been removed by the author.

  11. "If fiscal inflation does erupt, containing it will be difficult . . . with the debt-to-GDP ratio above 100%"

    Agreed! Since 1800, 51 out of 52 countries with gross government debt greater than 130% have defaulted, either through restructuring, devaluation, high inflation or outright default. The US crossed that threshold in 2020 and thus seems highly likely to default through sustained high inflation.

    See pages 3 and 6 of this letter:

  12. Consider a more intricate alternative theory. Covid shocks induce a switch of demand from services to goods. The price of goods rises, but the price of services may not decline as easily as suggested. A goods producer facing slack demand will see inventories build up and need to reduce prices to get rid of excess inventory. But a service producer facing slack demand does not see a corresponding inventory build up. An Uber driver who finds no passengers does not find the next day that his ride inventory has built up. Disney and American Airlines do not accumulate an inventory of unsold tickets. Further, the main cost of most services is labor, and as noted by Dr. Cochrane, labor is becoming more expensive and difficult to find due to government policy. If price is near marginal cost, service providers may even increase prices to compensate for higher labor costs and make due with far lower demand. Growth in money supply is not required, but the theory predicts we see inflation in the goods sector, lower real GDP in the services sector, overall nominal GDP is unchanged, overall real GDP is lower, and higher overall inflation.

    1. I find this to be a compelling argument. Would be interested to hear the author's rebuttal.

    2. Downwardly sticky prices in service industries. Possible, but it means those prices decline eventually, and inflation does not just go away, inflation reverses and we get back to the old price level. That does not seem to be in the cards. BTW, I get 1000 words and used 998, so I can't treat every possible complication.

    3. Sorry, but I think the logic here is flawed.
      "But a service producer facing slack demand does not see a corresponding inventory build up."
      Actually service providers will find their are "building up" time sitting idle. A consultant will find less billable hours, an uber driver sitting in the car idle, etc. Prices may or may not be sticky, but the reason they are or are not is independent of whether it's a physical good or a service.
      "Uber driver who finds no passengers does not find the next day that his ride inventory has built up. Disney and American Airlines do not accumulate an inventory of unsold tickets."
      Again, this might be an argument for how a supply chain is affected (bull whip effect, etc.) but the behavioral/economic implication is no different. The marginal cost of additional hours or the same seat on a flight remaining empty day after day increases with time.
      IMHO, good economic modelling recognizes that individuals only value services, not goods themselves (they value the services derived from the physical goods), and we should think in terms of marginal benefits and marginal costs. There's no fundamental economic analysis difference between the service industry and the goods industry. Once you recognize the "goods" themselves are fundamentally again just inputs to a technology providing the services individuals are actually deriving utility from then it's exactly the same in every regard. The empty seat on a place is the good unsold and so is the hamburger cooked but not sold. Both are inputs to a final service, unsold. The burger has a higher depreciation rate of course (and other "goods" might have lower depreciation rates), but otherwise the same.

    4. Travel services clearly saw downward price adjustments in the face of less demand. I don't think inventory levels is the driver of firm behavior - I think demand is.


Comments are welcome. Keep it short, polite, and on topic.

Thanks to a few abusers I am now moderating comments. I welcome thoughtful disagreement. I will block comments with insulting or abusive language. I'm also blocking totally inane comments. Try to make some sense. I am much more likely to allow critical comments if you have the honesty and courage to use your real name.