I have been tracking the monthly inflation rate (I use http://inflationdata.com/inflation/inflation_rate/historicalinflation.aspx . not because I particularly like his economic theories but his data is accurate) and I noticed something odd.
Every month so far in 2018 the inflation rate has increased. I can’t find any other year going back to the 70’s where that has happened. Has anyone else commented on this? Is it possibly significant? Or since it has never happened before should it be dismissed as a fluke? Or is it simply the byproduct that given inflation is still low, it is harder to go down than go up so increases are more likely.
And of course we have the rest of the year to go, but there hasn’t been a year with such a long stretch of increasing inflation (7 months) before.
Looking at your data I see non-decreasing inflation rates in 8 consecutive months or more for periods ending Mid-1956 (14 months), Mid-1934, Mid-1923, Mid-1925, Mid-1979, Late-1933, Late-1941, Mid-1989, Early-1947, Late-1939, Mid-1962 — compared with only six months in 2018 to date.
I write “non-decreasing” because, as you’ll see examining the table data, as recently as 1964 wherever these data came from didn’t recognize small changes. Eleven of the months in 1955-1956 had inflation rates of exactly 0.37% or -.37%. With this sort of data, “non-decreasing” rather than “increasing” is your only hope to compare apples with apples.
As OP points out, inflation has room to rise: there were seven months of deflation during 2009. Still, there was essentially no inflation in 2015, while now it’s nearly 3%.
A more interesting question is: Why the increasing inflation, and will this trend continue? This would detour the thread into politics but I think mentioning a recent tweet from the Commander-in-Chief is in order:
A baseline consideration for how interesting this might be is to imagine that inflation is a random coin flip (obviously, it’s not, although if you’re measuring it to hundredths of a percent, it might kinda behave that way a lot). If it were, you should expect a string of 7 months of increasing inflation in a row to happen 1 in 2^6 of the time, or 1 in 64 7-month strings. If you further restrict it to having all 7 of those months in the same calendar year (as opposed to a string that straddles more than one), then you only have 5 possible starting months per year, so you should expect to see such a string slightly less often than once a decade.
So, it happening just once in a ~45 year span strikes me as: a slight outlier, but not that interesting.
A more sophisticated model of the inflation rate might make it more interesting, but probably not. For example, the ways in which inflation is not like a coinflip make it tend to revert to a mean (both because noise in the measurement is random and because policy changes take place when it strays too far from what the central bankers want it to be), which makes the actual likelihood a bit less often.
Also, realize that you’re looking at a specific pattern and noticing something (apparently) interesting about it. If you saw a different pattern, you might think that pattern was interesting. Like, you could just as easily noticed that, say, inflation alternated increasing and decreasing in every year in 1987, and that didn’t happen in any other year since the 1930s (I made that up, don’t go looking for it). Which is true (it’s not) but so what? If you look at enough random noisy data you’ll see lots of things that aren’t interesting.
2011 fits the pattern identified, but for one month with a 0.01 percentage point drop (and the absolute increase was much greater in 2011 than in 2018). Indeed, the longer period Jun. 2010 - Sep. 2011 comes close as well.
Thanks all.
I can quibble with some of the identified years, but the main point is that this is likely to be a statistical fluke-random variation just happened to result in this string. So while it is interesting primarily because it is so recent-I doubt I would have noticed if the data were 5 years old-it doesn’t amount to anything significant.
I personally don’t look at economic data prior to 1965 for lessons or ideas about today’s world. That far back the economic world was so different I don’t think it can demonstrate anything about how the world works today. People have the same goals and pressures, but the external forces, government business labor supply etc, were so different and so strong that I personally don’t think one can compare the two ages. Others of course may differ with me on that.