Tuesday, April 13, 2021

Inflation levels

 March inflation is up. The CEA delivered a historic tweetstorm. It starts with 

temporary factors: base effects, supply chain disruptions, and pent-up demand, especially for services

I'm glad for once to have nailed a forecast: That Fed and Administration's first response to inflation would be to invoke "temporary" factors, just as in the 1970s.  We'll see how that pans out. 

The CEA goes on to "base effects,"

In the near-term, we and other analysts expect to see “base-effects” in annual inflation measures. Such effects occur when the base, or initial month, of a growth rate is unusually low or high..

This unusually large price decrease early in the pandemic made April 2020 a low base. 

Since this is about the past, we can say something more definite. Yes, if you start from a low base, you can see a lot of growth. To get around the arbitrariness, let's look at price levels. Here is the recent CPI (blue) and CPI less food and energy (red). These are the levels of the CPI -- conceptually how many dollars it takes today ($271) to buy what $100 bought in 1983.  









The last few months uptick is clearly visible. You can find your own "base month" by drawing a line. Yes, a line from last April to today shows an unusually higher slope, because last April was unusually low. 

But to the extent that we're just seeing "reflation," a return of prices to normal after a steep covid-induced recession, the graph suggests that was over last summer. "Reflation" was over by September. Draw your own trends -- that's why I left some history in. 

The inflation data is also seasonally adjusted. This is a big black box, involving a seven-year two-sided weighted moving average. It's one big reason data keep getting revised. For annual growth rates, it makes  sense to also look at the real data: 


This is non-seasonally adjusted CPI (blue) and CPI less food and energy (red). If anything the story, reflation was over last fall and this is something new is clearer. But, as 2019 reminds us in both graphs, stuff happens. 

BTW, here is the CPI through the last recession 


Look really hard to see the trend down in the red, core CPI, line around 2010. Befitting a recession driven by a financial crisis, not a supply-side pandemic shutdown, and prolonged (in my view) by counterproductive policies, we see the price level stay permanently below the previous trend.   


20 comments:

  1. How do demographic trends influence your thoughts on inflation trends?

    The shift of a large cohort from older workers to retirees (naively) suggests a fall in investment asset prices relative to consumption prices. Older workers purchase a lot of investment assets, whereas retirees sell investment assets. Lower demand and higher supply suggests a lower price of investment assets relative to consumption goods and services. At the same time, fewer workers (especially fewer highly experienced workers) suggests that there will be less consumption goods and services. Lower labor supply suggests higher prices for consumption goods and services (net demand changes from retirement are less obvious)(also, the forces on consumption stuff is less direct than the forces on investment assets).

    There seems to be two extremes for this convergence of investment and consumption: a fall in investment prices, or a rise in consumption prices (inflation). How much of this convergence is likely to be covered by inflation? How does this crisis (and possibly other crises) influence the distribution of price changes?

    ReplyDelete
  2. I wonder if we're obtaining a cleaner, less convoluted comparison of the price level as the consumption bundle reverts towards the pre-pandemic bundle. Might inflation have been understated for awhile now in official sources due to stale weights on various goods and services?

    The unprecedented change in the consumption bundle due is a huge complication. Cheap airline tickets aren't terribly relevant when you're unable to use them. Urban living is temporarily a much lower quality good than it used to be.

    For example Cavallo showed higher inflation than reported officially when using alternative weights based upon actual pandemic period spending from credit and debit transactions.

    ReplyDelete
  3. Can you elucidate if you can on the transmissions mechanisms on how inflation will rise to dangerous levels.
    I would be truly interested in understanding how a wage explosion is going to occur as in the 1970s.

    ReplyDelete
  4. Perhaps run the same charts, but use the PCE core.

    A lot of measured inflation in the US is due to restrictive housing policies.

    Even with a gold standard, if the supply of a major sector is restricted, and relatively declines, you will have inflation as measured.

    ReplyDelete
  5. Excellent post! as usual

    Silly question (to John, or commentators): is this CPI metric adjusted by how much of household income is dedicated to each good in the homogeneous consumption basket? To be more specific, if bread prices increase 30\% but my rent stays the same, I don't really care much about that inflation. But if rent increases 5\% and bread prices don't change, I will be really upset.

    Thank you!

    ReplyDelete
    Replies
    1. Yes, you can see the specific weightings here:https://www.bls.gov/cpi/tables/relative-importance/2016.pdf

      Housing makes up 42% of the basket while bread is 0.21%.

      Delete
  6. From March 1, 2008 through March 1, 2021, the trend has been ~EXPON((0.0160909/yr)·Δt) in the seasonally-adjusted series. Standard error of the estimate of the trend rate is 0.000187226/yr (F=7386.317411, DOF=155). Qualitatively, the 2007-2009 recession broke the prior trend, as you mention in your last paragraph.

    The question that needs to asked and answered is: Will the current pandemic-recession cause a break in trend, possibly to a higher trend rate? Conversely, will we experience an increasing rate of decline in the purchasing power of the dollar going forward? And, if so, how will the FOMC reaction and, perhaps more importantly, when will it react?

    It's not an academic question.

    ReplyDelete
    Replies
    1. This comment has been removed by the author.

      Delete
    2. Old Eagle Eye,

      "And, if so, how will the FOMC react and, perhaps more importantly, when will it react?"

      My own suspicion is that they will pull out the Seargent Shultz defense ("I know nothing, nothing...).

      During the last "major" outbreak of inflation, the FOMC went one way:
      Fed Funds Rate
      August 2007 - 5%
      June 2008 - 0.2%

      CPI Inflation Rate
      August 2007 - 2.69%
      June 2008 - 5.50%

      Meanwhile over the same time period corporate bonds went the other way:
      August 2007 (6.59%)
      June 2008 (7.04%)
      October 2008 (9.54%)

      I would expect more of the same.

      Delete
    3. FRestly,
      In response to the interesting data points in your posting, a quick excursion was made to the St. Louis Fed's FRED econ data charting website to garner a broader view of the time series. This is what an 'eagle eye' view of the time series data suggests:

      The time series (daily) for the effective Fed-Funds Rate, the constant maturity rates for 3-month T-bills, 10-year T-note, and 30-year T-bond, and the trimmed mean PCE inflation rate (monthly) show the effect that the Fed-Funds Rate ramp up and ramp down policy had prior to the last (2008-9) and current (2020-1) recessions. The recessions appear when the Fed-Funds Rate rises to/above the 10-year and 30-year constant maturity rates (yield curve inverts). The 3-month constant maturity T-bill rate follows the Fed-Funds Rate with some variation.

      The graph (URL below) shows inflation lagging the Fed-Funds Rate movements. This would explain in part the data points that you have chosen to show.

      Contrary to neo-Keynesian theory, the inflation rate (PCE) data does not support the theoretical effect of a varying Fed-Funds Rate on the inflation rate. Perhaps, PCE is not the right type of inflation rate?

      Both the 10-year and the 30-year constant maturity interest rate time series data appear to follow paths different from the Fed-Funds Rate. Your observation of the 10-year yield conforms to this notion. When the 10-year and 30-year interest rates have dropped, it is usually at a point of stress to the financial system: e.g., 12/29/08, and 3/26/20-4/7/20.

      Inference? The FOMC has little in the way of policy influence on the rate of PCE inflation. 'Managing expectations' is not viable as a policy tool, though it might appear to be at times. The FOMC has used the Fed-Funds Rate to control speculative activity in the markets, but that works best through the lending rate (prime, and broker call rates). There is no data indicator for determining when the FOMC will alter its Fed-Funds Rate; when it does alter the rate, it does so slowly and step-wise. There is no evidence that it does so in response to PEC inflation rates. Perhaps it looks at employment and production capacity state variables.

      URL: https://fred.stlouisfed.org/graph/?g=D8FK
      Date range: 2003-04-20 to 2021-04-14
      Plots five sets of data graphically:

      (1) Effective Federal Funds Rate (DFF)
      Source: Board of Governors of the Federal Reserve System (US)
      Release: H.15 Selected Interest Rates
      Units: Percent, Not Seasonally Adjusted
      Frequency: Daily, 7-Day

      (2) 10-Year Treasury Constant Maturity Rate (DGS10)
      Source: Board of Governors of the Federal Reserve System (US)
      Release: H.15 Selected Interest Rates
      Units: Percent, Not Seasonally Adjusted
      Frequency: Daily

      (3) 3-Month Treasury Constant Maturity Rate (DGS3MO)
      Source: Board of Governors of the Federal Reserve System (US)
      Release: H.15 Selected Interest Rates
      Units: Percent, Not Seasonally Adjusted
      Frequency: Daily

      (4) 30-Year Treasury Constant Maturity Rate (DGS30)
      Source: Board of Governors of the Federal Reserve System (US)
      Release: H.15 Selected Interest Rates
      Units: Percent, Not Seasonally Adjusted
      Frequency: Daily

      (5) Trimmed Mean PCE Inflation Rate (PCETRIM12M159SFRBDAL)
      Source: Federal Reserve Bank of Dallas
      Release: Trimmed Mean PCE Inflation Rate
      Units: Percent Change from Year Ago, Seasonally Adjusted
      Frequency: Monthly

      Delete
  7. I have just one question: How do retail investor profit from this?

    I always try and profit from leftists' irrationality. It has been going really well. I bought the stock market at bargain prices in April induced by their lockdowns, for example.

    But inflation I don't know how to trade. Should we buy indebted companies, maybe? Real estate? Bitcoin, gold, etc?

    Assume that inflation will increase... A LOT. And assume that this is a surprise. Any advice on profitable trades?

    ReplyDelete
    Replies
    1. "Any advice on profitable trades?"

      Get off your desk chair, go outside, produce some good that other people can use and trade it for money.

      Delete
    2. This comment has been removed by the author.

      Delete
    3. Not that I think that anyone that matters would not know this... but I am obviously talking about securities trading. ;)

      Delete
  8. John,

    Please do a post on yield curve control (YCC). I am worried that when longer term rates start rising and thus putting pressure on Federal government's fiscal imbalances, the left will put strong pressure on the Fed to do YCC (because it wants to continue with deficit-financed spending). This seems like the recipe for monetary expansion and inflation. What are your thoughts on this?

    ReplyDelete
  9. I just want to add that at the end of March, the Fed discontinued reporting all money supply measures. And you can't "roll" you own. The Fed also discontinued publishing data on currency, demand deposits, savings accounts, small and large time deposits, and retail and institutional money market funds.

    Curious how this happened right after the Biden Administration got rolling.

    Could it be that Biden and the Fed don't want us to know how rapidly the money supply is rising?

    ReplyDelete
    Replies
    1. This comment has been removed by the author.

      Delete
    2. This comment has been removed by the author.

      Delete
    3. Anonymous,

      You can still find most of these things (demand deposits, money market funds, etc.) inside the Federal Reserve's quarterly Z1 flow of funds report.

      The Fed has stopped publishing monetary aggregate measures for quite some time.

      https://www.federalreserve.gov/releases/z1/default.htm

      Delete

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.