But when I went back and tried to calculate some numbers myself, I did not see recovery before late 1944....the "recovery" [from 1929] was short-lived.... by most measures, yes, the market had recovered by the end of 1936.... But only just barely, and in 1937 it crashed again. That was a decline of -49.93%, so it was just about the same as 2008-2009. Apparently the second crash escapes notice because the 1929 crash was so much worse.
Using the Ibbotson SBBI 2015 Classic Yearbook as an objective source--or, at any rate, not my own judgement--their presentation of the data is: (p. 172, selected lines from table 13-4, "Largest Declines in U.S. Stock Market History)
Peak Aug. 1929, Trough May 1932, Decline 79.00%, Recovery Nov. 1936
Peak Feb. 1937, Trough Mar. 1938, Decline 49.93%, Recovery Feb. 1945
Here's how I got my numbers. I used the month-by-month numbers from the 2015 Ibbotson SBBI Classic Yearbook, p 196, table A-1, "Large-Capitalization Stocks, Total Return." For anyone interested in checking my work, the monthly returns for 1929, starting in January, were:
0.0583, -0.0019, -0.0012, 0.0176, -0.0362, 0.1140, 0.0471, 0.1028, -0.0476, -0.1973, -0.1246, 0.0282
I assumed $10,000 at the end of 8/1929, and then began calculating cumulative growth values from the total return numbers starting with -0.0476.
I was surprised by the results, so I cross-checked by getting data from a different source, Shiller's "Irrational Exuberance" data, available online at his website. The SBBI data might be more reliable because I got them straight from the source. For the Shiller data, I had to derive them by making a calculation from separate price and dividend values; there also might be a date-alignment issue since SBBI data is month-end while I think Shiller's is month-average. Anyway, the two growth charts are reasonably close.
Here's what I got:
As you see, an investment of $10,000 would not have gotten back to even, or even close to it, at the end of 1937. The local maximum, using SBBI data, was $8,543.13 on 2/28/1937; using the Shiller data, $8,808.61.
A $10,000 investment stock market was not back to even until
1/31/1945, $10,063.01, using the SBBI data.
11/31/1944, $10,099.01, using the Shiller data.
In either case, we have at least a 15-year-long bear market, and at least 13 years from bottom to recovery.
In reality, for most purposes it doesn't matter much whether the peak in 1937, after two of the best years in stock market history, was just above or just below the starting point. Nevertheless, it's a detail worth pinning down.
So, did I screw up somehow?
Was someone who invested at the peak in 1929 back to even by 1937, or not until 1945?