A quick note on inflation

The TIPS spreads are a useful indicator of inflation expectations.  But they are not perfect.  The spread can be distorted by lots of factors, including changes in the risk premium and sudden swings in commodity prices (the TIPS adjustment occurs with a lag, so actual changes in commodity prices show up as expected changes in the CPI.)

Today, the 5-year TIPS spread is 1.77%.  Because it is based on the CPI, it corresponds to roughly 1.5% inflation on the PCE, which is the inflation rate that the Fed actually targets (at 2%.)

So if you believe TIPS spreads accurately measure inflation expectations, you are forced to believe that investors currently expect inflation to fall well below the Fed’s target over the next 5 years.

Over the past 12 months the PCE inflation rate has run at 1.37%, so it’s likely that investors expect a pick-up in inflation over the next 5 years.  But that was probably true before the election as well.  The core PCE is running at 1.7%, and most experts believe the core is a better indicator of future inflation, as it’s not effected by transitory swings in commodity prices.  It’s not that food and energy “don’t matter”, of course they matter a lot for consumer living standards, the point is that their relative prices are roughly a random walk, and hence the core is a better tool for forecasting inflation.

To summarize, the recent headlines about “inflation coming because of Trump” are only half true.  Yes, the TIPS spread has widened since the election, and I think it’s very possible that the increase was in part caused by the election.  That suggests that monetary offset is not perfect, not 100%.  On the other hand, in absolute terms the magnitudes are still quite low.  I still think we are looking at a 2% inflation economy, or slightly below, as far as the eye can see.  If it has increased a bit, we don’t know why.  Is it fiscal stimulus, or the expectation that Trump will appoint another Arthur Burns in 2018, or both?  I simply don’t know.  TIPS spreads often move around for reasons that are poorly understood.

FWIW, I’ve argued for years that 1.8% inflation is the new normal, not because the Fed wants it, rather because the policy that the Fed thinks will give them 2% inflation will probably actually give them about 1.8%, on average. Two percent during good times and 1% to 1.5% during bad times.  The Fed always tends to assume good times ahead. The recent moves in TIPS spreads are potentially important, but I’d need more evidence to move me off my 1.8% forecast.

PS. If you use 10-year TIPS, the spread is about 15 basis points higher, still pointing to sub-2% inflation going forward.

PPS.  As always, a NGDP futures market would add useful information.


Tags:

 
 
 

13 Responses to “A quick note on inflation”

  1. Gravatar of Joe Leider Joe Leider
    14. November 2016 at 08:59

    How can increased uncertainty drive inflation expectations? Trump seems like the most unpredictable President in quite some time. I’ve felt it in my own buying behavior. With things so up in the air and the probability of a nuclear war just a little higher, I might as well spend a little more now and enjoy myself. In aggregate, could that increase velocity and thus inflation expectations?

  2. Gravatar of ssumner ssumner
    14. November 2016 at 09:00

    Joe, That’s possible, although monetary offset can limit the effects.

  3. Gravatar of ant1900 ant1900
    14. November 2016 at 09:21

    What’s the case for Trump not moving the Fed in a more dovish direction, either through persuasion or appointments. That could explain an increase in inflation expectations and preserves monetary offset.

  4. Gravatar of Fran Fran
    14. November 2016 at 12:52

    I completely agree that we are more likely to experience continued low inflation rather than some sudden increase. Infrastructure spending won’t do the job of increasing inflation. Japan is an obvious counter example of the effects of monetary offset. Yet the media and some commentators seem to forget this entirely. At the end, one risk we face is that we might end up having a more hawkish Fed that goes too far with the offset, with corresponding economic consequences.
    My question: wouldn’t supply-side policies ease price pressures, leading cp to lower inflation and lower bond yields? To me, the observed increase in bond yields remains mysterious.

  5. Gravatar of ssumner ssumner
    15. November 2016 at 06:51

    ant1900 and Fran, I think the strongest argument goes as follows. The Fed normally aims for 2%, but falls short in recessions. So inflation (post-2007) averages a bit under 2%. Trump’s supply-side reforms will make a recession less likely. So inflation expectations rise, but remain under 2%.

  6. Gravatar of Michael Michael
    15. November 2016 at 12:39

    Professor,
    you complained (in another post) that this one got ignored. Inguess you trained us too well ? that a) inflation doesn’t matter b) regimes matter and anyway, c) the big picture matters, not a weak signal drowning in market noise

    That said, I find your derivation of 1.8% average inflation quite ingenious, and I think it explains the facts quite well. It’s a bit like Friedman’s pulling of a string

  7. Gravatar of Ricardo Ricardo
    15. November 2016 at 17:46

    All the news stories seem to imply the sharp rise in bond yields (mostly the inflation component) are due to expectations of increased govt spending (or govt deficits). Even stories about “bond vigilantes” are back in the headlines.

    I realize you have issues believing that since you believe the Fed is firmly committed to 2%. And, I do realize that many times it is “price that dictates news” … meaning that the news writers make up a story to fit the price movement.

    Still, to me, the best correlation is simply with the likely-hood and size of “infrastructure” spending. I know you don’t agree, however. It’ll take more news to clear that up.

  8. Gravatar of ssumner ssumner
    15. November 2016 at 18:54

    Thanks Michael.

    Ricardo, I have on open mind on that. I see monetary offset in the same was as the EMH, a good approximation of reality but probably not 100% true. Why am I pushing back a bit here? Two reasons:

    1. Inflation forecasts from TIPS are still well under 2%

    2. TIPS spreads are known to react to other things besides inflation expectations.

    I agree that TIPS provide SOME evidence that inflation expectations have risen, but the rise is still fairly modest, and we’re still not sure why. Another possibility is that 2% is still the modal forecast, but there is now less risk that inflation will fall short, because Trump’s actions will make a recession less likely (partly for supply-side reasons). Or maybe the Fed won’t do 100% offset, that’s possible. I’m going to be interested in how the consensus forecast of private economists changes, and also how the Fed’s forecast changes. Those two are usually pretty close.

    Also note that the dollar is appreciating, something that usually holds inflation down.

  9. Gravatar of Ricardo Ricardo
    15. November 2016 at 20:19

    I just solved the zero-bound problem! Next time there are deflation fears, just suggest giving Donald Trump complete power over Fed monetary policy. Haha!

  10. Gravatar of B Cole B Cole
    16. November 2016 at 06:13

    Trump’s infrastructure spending, even if fully implemented, will run about $100 billion a year.

    National Security spending is running at about $1 trillion per year.

    The cost of the Afghanistan and Iraq Wars was somewhere around $6 trillion.

    GOP’ers are blubbering about “re-building” the military. You know, because we only spent $16 trillion in the last 16 years on national security (excluding wars).

    The bond market cares about Trump’s infrastructure spending? Is innumeracy the defining trait macroeconomists and bond traders?

  11. Gravatar of ssumner ssumner
    16. November 2016 at 07:13

    Ben, You said:

    “National Security spending is running at about $1 trillion per year.”

    No it isn’t.

  12. Gravatar of Ricardo Ricardo
    16. November 2016 at 16:00

    OK, +1 for your “Arthur Burns” scenario.

    Today there were a couple of stories very similar to this:
    “Trump could cause a train wreck at the Fed”

    https://www.washingtonpost.com/opinions/global-opinions/trump-could-cause-a-train-wreck-at-the-fed/2016/11/15/61b48f60-aabc-11e6-a31b-4b6397e625d0_story.html?tid=a_inl

  13. Gravatar of David Englund David Englund
    18. November 2016 at 08:00

    Correct me if I’m wrong, but I thought inflation was caused by wage pressures. IE companies are forced to raise prices to cover wage increases leading to higher inflation. I don’t see wages rising appreciable anytime soon.

Leave a Reply