Charts are everywhere and, especially in the finance world, analysts will make decisions based on charts more than they will on words or paragraphs. But charts can be misleading, and smarter analysts can use misleading charts to lead the less experienced analyst astray.
This isn’t an accusation of dishonesty; one can use real, authentic data and accurate information in a completely honest way that still leads people to make bad decisions. That’s because the presentation of the data and the interpretation of it depend on context and, most crucially, understanding what the chart means.
To explain this, let’s use the example of the CPI. The Consumer Price Index is recorded monthly by the Bureau of Labor Statistics, and it remains both one of the most important and one of the most widely used macroeconomic indicators in the financial world. Since it measures the costs that consumers face in the marketplace, it obviously has a lot of importance for understanding consumption and inflation.
Now let’s take a look at a chart of the CPI and try to interpret it.
The orange line here looks alarming–rising at the start of 2020, and suddenly rising more and more sharply in 2021 to a nearly logarithmic clip. This is the CPI itself, and it looks like it’s simply getting out of control. The orange line is telling us that not only is inflation red hot, but it’s overheating and we’re flirting with a hyperinflationary disaster.
The light-blue line tells us a different story. As a month-over-month indicator, it shows a pretty steady pattern of sub 1% growth and, most crucially, a downward trend in the last two months. That suggests inflation is starting to cool down.
The purple line, on the other hand, is a year-over-year indicator, and it too shows a sharp growth starting in March 2021, but it has crested and appears to be at the start of a decline. That tells us inflation started to heat up a few months ago and is now starting to cool down.
These are all measuring the same thing–the CPI–but they are telling us three different stories. Which do we believe?
Well, the orange line is just the CPI itself–that is a line that has been going up for over a century, with occasional dips during recessions and market panics. That line by itself really tells us little, especially for such a short-term time horizon.
The month-over-month data will get us short-term trends, which is good for identifying a sudden shock (this data was particularly important during the start of the pandemic), but its short-term limits also mean it is going to introduce a lot of noise–even if you seasonally adjust month-over-month data, unusual events (like an economy reopening, or a jump in vaccination rates) will make it hard to use this data to identify a long-term trend.
Year-over-year is in many cases the gold standard of economic metrics. This avoids seasonal variations (because you’re choosing the same season a year apart) while also giving you a long enough time horizon to identify a trend. While not always useful, the year-over-year trend is typically preferred over month-over-month data to identify the shape of a long-term trend.
And that’s the line that most market participants have been using when it comes to our current inflation situation. A sharp jump in March and April 2021, compared to a year ago when the U.S. was in lockdown and the country was widely in a panic about a little known new disease, makes a lot of sense–and it also means inflation is not so much a fundamental concern as the data would look prima facie. A slowdown in the YoY growth rate in June and July also makes sense intuitively (we’re comparing now to the start of the reopened economy, but still pre-vaccine when many Americans remained wary of going out and spending).
This narrative also gives us a reasonable model for future expectations: A further slowdown in YoY CPI growth as we compare against a more open economy is likely, with a significant decline in YoY data to start in January 2022 as we compare against the start of the vaccination drive. March to May 2022 will have very tough comparables a year ago, indicating a YoY decline in the CPI is likely then.
We can draw a chart in our minds of how this will look–and it looks like an S curve, as the sharp rise in early 2021 falls in mid-2021 and goes below the start in early 2022. That then can be our model for the future–and we can invest accordingly, either going bullish or bearish if it overshoots accordingly.