The ratio of total listed market capitalisation to nominal gross domestic product has one great virtue. Its denominator is not an accounting number, so no write-off, no depreciation policy and no buyback can touch it. That is also the source of its defect: the denominator measures the whole economy, and the numerator prices only the listed part of it.
The defect is severe in India. Agriculture is a large share of value added and essentially unlisted. The unorganised sector is larger still. On an illustrative but plausible assumption that listed companies account for 35 percent of value added, the raw ratio of 127.78 percent becomes 365.08 percent when measured against the output the listed sector actually produces. The raw ratio has not become wrong; it has become a statement about two things at once, the valuation of the listed sector and the share of the economy that is listed, and those move independently.
Advertisement
Why can't you compare market cap to GDP across countries?
The defect runs the other way in the United States. Large index constituents earn a substantial share of revenue abroad, so the numerator prices cash flows the domestic denominator does not contain. On an illustrative 30 percent foreign-revenue share, the raw ratio of 203.28 percent falls to 142.30 percent once the numerator is restricted to domestic claims, which is identical to grossing the denominator up to 43.57 trillion.
Put the two adjustments side by side and the naive cross-market reading reverses. Raw, India at 127.78 percent looks 75.5 percentage points cheaper than the United States at 203.28 percent. Adjusted, India at 365.08 percent looks 222.8 percentage points dearer than the United States at 142.30 percent. Nothing about either market changed between those two sentences. The sign flip is entirely a property of the denominator you chose, which is why the ratio is usable against a single market's own history and useless as a cross-country comparison of levels.
How do you use market cap to GDP within one market?
Read against its own history the ratio behaves well. On the illustrative Indian series it sits at 127.78 percent against a ten-year mean of 94.57 percent, in the 80th percentile of its own prior decade. The signal there is not "India is expensive relative to America". The signal is "the listed sector's capitalisation has grown a lot faster than the economy for several years, and either the listed share of the economy is rising, which is a real and observable process, or the multiple is".
Distinguishing those two possibilities is a task with an answer. Count new listings and their capitalisation, and track the listed sector's revenue against nominal output. When the ratio rises because the listed universe is expanding, aggregate listed revenue rises with it. When it rises because the multiple is expanding, listed revenue does not move and the multiple does.