Our ‘Airing of Grievances’ With Financial Commentary

It is December 23, and you know what that means: Well, yes, two days to Christmas and Hanukkah! Yes, yes, three until Kwanzaa. And, yes, yes, yes, only eight days remain in 2024. But there is something more—something that deeply resonates with us huge fans of most things 1990s: December 23 is Festivus. And here we intend to bring you a traditional Festivus “airing of grievances” in which we share our chief gripes with things we have read in the financial press this year. It is a way we aim to enter the new year unburdened by what has been, if you will forgive us invoking Vice President Kamala Harris.

For the uninitiated, the story of Festivus starts in 1966, when author Daniel O’Keefe dreamed up a holiday repudiating what he saw as the holiday season’s increased commercialization. He launched a family tradition involving nailing a clock in a bag to the wall, eating a turkey or ham and then “airing the grievances”—telling friends and family all the things they did to annoy you that year.

The popularity took off in December 1997, when his son Dan—a television writer—worked on a now-famous episode of Seinfeld that turned the clock-in-bag into an aluminum “Festivus pole,” the dinner meat to meatloaf … but kept the grievance-airing. As the legendary Jerry Stiller kicked it off, “I got a lot of problems with you people—and now you’re going to hear about it!”

In that noble spirit, we have a lot of problems with some of the financial punditry’s work … and you can read all about it!

State Market Movement in Percentages!

When markets hit turbulence last Wednesday, headlines predictably shrieked “DOW FALLS 1,100 POINTS!” Earlier, on the upside, many hyped bitcoin breaching $100,000. We have long had a bone to pick with “point” or value metrics. The reason: Is that big? Is it supposed to be big? Without context, you can’t know. And this problem is getting worse with time.

Yes, the Dow’s falling -1,100 points is a fairly big daily decline … now … as we will explain.^[i] (And, yes, the Dow is a horribly flawed, price-weighted index of just 30 US-only stocks, a terrible means of measuring market movement—stay with us anyway.) But quoting it in points makes it sound really ginormous. We mean, 1987’s Black Monday was -508 points! And therein lies the problem with point-movement reporting: Black Monday’s -508 points were a massive -22.6% decline! Wednesday’s -1,100? -2.6%. That is not likely to garner it any nicknames, considering the gauge has risen or fallen that much 1,055 times (or 2.8% of trading days) since the Dow’s 1885 birth.^[ii] A point decline the magnitude of Black Monday is now just -1.20%, a magnitude seen 6,251 times in history (16.3% of trading days). Stop thinking in points. Start thinking in percentages.

This problem is only going to get worse as index numbers get bigger. And it is a corollary to investors thinking about their portfolios’ daily dollar change, a flawed view. When your portfolio gets larger, the daily dollar change with market movement will likely be bigger, too. Expect it. Here again, the right measure is the percentage change, not the dollar, point or absolute figure.

Referring to Yield Spreads as “Borrowing Costs.”

For the past month, we have seen article after article claim that French borrowing costs jumped as the government collapsed. Always stated as fact, never with actual numbers, but it would give any reasonable reader the impression that French long-term interest rates, you know, rose, surged, spiked … you pick the action word, folks. Went up.

But what actually rose? Not French interest rates! It was the gap between French yields and German yields for the same maturity. French yields, in absolute terms, fell.^[iii] German yields fell more, so by that one very specific comparison, France became more expensive in relative terms. Relative to Germany. But the actual, absolute cost of French borrowing? It fell.

Conflating Percent and Percentage Points

Ok this is very finicky and math-adjacent, but it matters! We have lost track of how many times we have seen a sentence like this: “Widgets ‘R’ Us enjoyed earnings growth of 5.5% y/y, beating expectations by 2%.”

No.

If expectations were for 3.5% growth and results are 5.5%, results are not 2% higher. They are two percentage points higher.^[iv]

Similarly, when a bond yield rises from 1.3% to 2.4%, it doesn’t rise by 1.1%. It rises by 1.1 percentage points.

The difference between two percents is always, always, always measured in percentage points. This is not just us saying it. As Eurostat puts it, “The term percentage point is used when comparing two different percentages.” Or the UK Office for National Statistics, “A percentage point is the difference between percentages. A value of 10% falling by 1 percentage point becomes 9% (10% has 10 percentage points). A fall of 1% would result in a value of 9.9%.” Or just Merriam-Webster’s: “Percentage point: one hundredth of a whole.”

Relatedly, if we were supreme, and likely benevolent, leaders for a day, we would ban all use of percents-of-a-percent. Meaning, if GDP grew 2.0% in Q3 and 3.0% in Q4, we would bar all writers from saying Q4’s growth rate was 50% higher. They would be required to write that it was one percentage point higher. While mathematically true, the percent-of-percent figure is unclear and meaningless, and it is a way to turn small numbers into big ones. It is messy, and we don’t like it.

A grand irony here is this: Many who commit this fallacy cite basis points in their work (a basis point is one hundredth of a percentage point). Yet they seemingly fail to grasp that percentage point and percent aren’t synonymous and that can result in unclear communication.

A Singular Plural

Also in the realm of small and finicky but crucial, may we remind everyone that “data” is not the singular form of the noun, but rather a plural? The singular form is datum. Data refers to multiple datum, or, multiple data points. Which means … you must conjugate your verbs correctly!

The correct phrasing is not “this month’s inflation data was fine.” You can say the inflation report was fine. The headline inflation reading was fine. But if you choose to use “data,” you must say, “this month’s inflation data were fine.” Data are. Data were. Not “data shows wages growing at a solid clip,” but, “data show.”

There is only one exception where “data” is singular: if you are referring to the sentient android, Data, from Star Trek. But only Paul Krugman comes to mind as an economist with frequent Star Trek musings, and he is retiring as a columnist (hat tip, pour one out, all that), leaving Elisabeth to carry that particular nerdy torch.

And for all other uses of “data” … use plural forms of verbs!

Stating Unequivocally That Wage Growth Fuels Inflation

This one might seem like a philosophical difference, but we think it is more than that. You see, it isn’t a fact that wage growth leads to faster inflation. It is a theory. The thing about theories: You can test them against available data. When you test the history of wage growth against the history of inflation, you will see wages lag inflation. They do not lead it. They are an after effect. Milton Friedman explained the reasons for this in a famous 1960s speech, making the point that businesses factor in recent inflation when setting wages. And nearly 60 years’ worth of data since then show he is correct—including the past three years.

So is it too much to ask for some more careful wording? As in, when stating a theory, tell readers it is a theory, a school of thought. Do not state it as fact if it is not a fact. Do not state as fact anything you haven’t verified with available data. Otherwise, your 8^th grade English teacher will flunk you for making unsupported generalizations.

Relatedly: “Preferred” Doesn’t Equal “Targeted”

The inflation over the last few years has given rise (ha) to another sticky (ha) trend in financial commentary: In discussing almost any inflation measure, coverage compares the present year-over-year rate to central banks’ common, if unexplained, target of 2%. But there is a problem common to it all: The articles play loose and fast with the targeted measure, devaluing the analysis (ha).

Every major nation has more than one inflation metric. In the eurozone, you might have a nationally tallied consumer price index (CPI) and various flavors that strip out certain product groups, like food, fuel and alcohol. There are other measures like producer price indexes (PPI) or harmonized consumer price indexes that facilitate cross-border comparisons. In America, we have CPI, PPI and the personal consumption expenditures (PCE) price index—and we have core measures of all three that strip out volatile food and fuel. In Canada, there is CPI and various “trimmed” and “median” measures that remove outliers—its version of core.

The 2% target is specific. It isn’t just a general, pick-your-own-target. In America, the Fed specifically targets headline PCE. Not core, which many call its “preferred gauge.” Headline. In Canada, it is total CPI—not trimmed or trimmed mean or whatever, which analysts frequently call the Bank of Canada’s “preferred measure.” The Bank of England? Headline CPI. ECB? Harmonized CPI—headline. Not core. None of them. As we have written before, we have no idea what the, um, fine folks who run the world’s central banks prefer. But we know exactly what they target. Start writing like it and stop giving us comparisons of “core,” “super core” or “services” inflation to the targeted rate. It is useless.

Playing Fast and Loose with EU, Eurozone and Europe

We don’t know who needs to hear this, but a small public service announcement: The EU is a political group of 27 member states scattered across Continental Europe, Scandinavia and Ireland. The eurozone refers to the 20 states that use the euro as their primary currency. And Europe is one of the seven continents.

Friends, you cannot use these interchangeably. Not every country in Europe is in the eurozone or EU. Saying “European stocks” when you mean “EU stocks,” “eurozone stocks” or “European Ex. UK stocks” is inaccurate. If you are referring only to the euro countries, say “eurozone” or, if you prefer, “euro area” or more technically, the Economic and Monetary Union (EMU), but that is a mouthful. If you are referring to EU countries, say EU. If you are referring to the general area between the Ural Mountains and Atlantic Ocean—including the UK and Ireland—say Europe. If you are referring to the contiguous landmass west of the Urals only, say “Continental Europe.” If you are referring to this contiguous landmass and Ireland, say “Europe Ex. UK.” If you are referring to anything that includes Switzerland, the UK or Norway, don’t use EU or eurozone. Check a map, a flowchart, a Venn Diagram, whatever. (Preferably Venn Diagram, of course, we are so with Vice President Harris on this one.) But please for the love of all things pretty and warm, use the right word for the right entity.

With that, we bid you adieu and happy holidays. We hope you and yours have a warm, bright holiday in which everyone refers to the right inflation target, cites change in percent or percentage points appropriately, knows their spreads from their yields and never, ever, conflates their European entities. A true dream of a holiday.

Our best wishes to you and yours, dear readers.