# Can the stock market grow faster than GDP indefinitely?

My understanding of conventional wisdom is that average long-term growth of advanced economies is expected to be significantly lower than the average long-term growth of the value of the largest firms in the stock market (e.g. the total value of the FTSE 100). I also accept Kenny LJ's argument that GDP (and the stock market) can grow exponentially, indefinitely.

But if the largest firms in the stock market, taken as a group, always grow faster, doesn't that suggest either

1. an unsustainable trend, or
2. a fundamental restructuring of the economy is in progress?

Is there any limit to how many times larger FTSE 100 firms could be than the output of the entire economy, for instance?

EDIT: To be completely clear, I am not using the FTSE as a proxy for the wider stock market or private sector more generally, so I'm pretty sure survivor (or selection) bias is nothing to do with the answer. I am literally asking about the size of the group of the largest $x$ companies listed today relative to the total economy.

• It is usual optimistic for those who embrace particular school of economic. Speculation and fraud is the most important formula most "growth support" attempt to sideline. As long as any of the stock market index player continue to show "too good to be truth" results against general industry odd without solid proof (if you recall various accounting fraud and ponzi like scheme), you can bet something is very wrong there. – mootmoot Apr 23 '18 at 12:40

There are trends that has allowed stock markets in advanced economies to grow faster than GDP for a long time:

1. Branching out abroad. This gives access to faster growing markets in developing countries. This trend will end when all countries are advanced economies.

2. Fewer private companies. This trend will end when most of GDP is generated by companies listed on the stock market.

3. Multiple Expansion. Reduced interest rates and increased money supply can reduce returns expectations and increases PE multiples. This trend has worked both ways through time but for the last 10-20 years it has generally acted in an expansive capacity.

In addition, there is a way for the stock market to generate a higher return without an increase in stock market value:

1. Dividends. Dividends increase stock market return without increasing its value. Though not really considered growth, dividends are often included when comparing stock market growth with GDP growth.
• Survivorship bias.

The Dow Jones Industrial Average (DJIA) was first published at 40.94 on 1896-May-26. It closed at 24,448.69 on 2018-Apr-23. Let's call that 122 years. That's a 5.4% annual compound growth rate — certainly faster than the US GDP growth rate over that 122-year period.

However, the list of constituent companies that go into the computation of the DJIA has changed over time. The initial list contained just 12 companies, of which only one (General Electric) still survives.†

Almost by definition, the stock market consists of only companies that haven't completely failed. And so the average company listed on a stock market is more successful than the economy's average company.

Even if we look at broader indices or even a country's entire stock market capitalization, we will find that the same is true (albeit to a lesser extent) — in most countries, over long periods of time, the stock market grows faster than the country's GDP.

(The above is just 20 minutes worth of "research". I'd be interested to know what estimates researchers have come up with to quantify such survivorship bias and explaining the gap between GDP and stock market growth.)

Related point:

• Selection bias.

The stock market is not a representative sample of the entire economy.

For example, it is perfectly possible that one neighborhood of a city consistently gets richer faster than the rest of the city, perhaps because ever-richer folks keep moving into that neighborhood.

It is likewise possible that the stock market (which does not exactly reflect the entire economy) consistently attracts ever-more-successful companies.

(This is just one possible reason for selection bias; there may be others.)

† The 12 companies initially listed on the DJIA were (source):

• American Cotton Oil Company
• American Sugar Company
• American Tobacco Company
• Chicago Gas Company
• Distilling & Cattle Feeding Company
• General Electric
• Laclede Gas Company
• North American Utility Company
• Tennessee Coal & Iron
• U.S. Leather Company (preferred)
• U.S. Rubber Company

Here's a brief article stating what happened to each of the above companies. It is probably safe to say that if the DJIA stuck to these 12 companies and no matter what method we use to measure the present-day value of these 12 companies, the DJIA would be much lower than its current value of 24,000+.

• Selection bias isn't the reason why stock market returns exceed GDP growth. If in a given year one company goes bust (say Tennessee Coal & Iron) and another lists on the stock market (say Google) and others grow/shrink relative to the country's GDP then the sum of these effects will be reflected in the GDP / stock market ratio. Absent dividends (as explained in my answer below), entry and exit from the index cannot prevent the index from being a greater and greater share of GDP, which is unsustainable in the long-run. – oli5679 Apr 28 '18 at 11:58
• General Electric was removed from the DJIA in June 2018. – Flux Jun 4 at 12:34

If survivor bias means that more successful firms more typically list on a public exchange and less successful firms do so less often, then it would be possible for "the stock market" to grow faster than GDP indefinitely.

Perhaps trivially, and ignoring survivor bias implicit in stock market indices per se: you might also assert some condition where the share of capital grows in the economy by, say, 0.000000000000001% per year, or any other sufficiently small number, for NPV of profits from economywide capital to perpetually increase faster than GDP. This would be more likely if taking sufficiently long periods of analysis - for example, it is unlikely to hold in each and every single quarter, but it could hold in each and every decade.

1. A stake in the companies whose shares you own, including income that they generate and reinvest in new internal projects.

2. Access to a stream of income generated by these companies, which they distribute to shareholders through dividends, share-buybacks ect.

As you suggest, it is impossible for the growth of 1 to significantly exceed the growth in GDP indefinitely. However, an individual's portfolio return is made up of the growth of 1 + the returns of 2.

For example, suppose Apple makes a profit equal to 5% of it's market value each year (accounting for inflation, and depreciation of current assets). They can (1) - reinvest these profits in the company - buying new equipment, office space, intellectual property ect. (2) - pay it out to their shareholders, who can then spend/invest this money somewhere different. If the economy grows at 2% / year, and Apple allocates 40% of these profits to internal investment, it's share of GDP will not rise. However, an investor that reinvests the additional dividend (equal to 3% of their investment) in other opportunities will see their share of the economy rise each year.

Indeed, in an uncertain world made up of inpatient investors, they can be offered be offered a positive real return in exchange for forgoing consumption in the present that may well be higher than their economy's long-run growth-rate.

This means that net owners of shares can acquire a greater and greater proportion of GDP as time goes on. Thomas Piketty finds some empirical evidence for this over the last 200 years, which he names 'R > G'.

https://en.wikipedia.org/wiki/Capital_in_the_Twenty-First_Century

• My understanding is that (1) and (2) are two ways to look at the same thing, and people ultimately only want shares of publicly listed companies for (2) because shareholder influence over the company is basically non-existent (except for exceptional circumstances). Thanks for the link to Piketty, although as I understand his argument he is saying the 'r > g' trend is not long-term sustainable, and leads over time to massive fundamental changes in the economy. Is that your interpretation? – Dan Apr 24 '18 at 10:34
• Added an example above. (1) and (2) are distinct and mutually exclusive. – oli5679 Apr 28 '18 at 11:42

Some great answers miss your point because they focus on indexes, whereas you are asking simply about the market value of the x largest companies relative to gdp, if there is a limit. Indices are plagued with selection bias, survivorship bias, dividend reinvestment, acquisitions.

My humble opinion is that barring companies owning the shares of other public companies directly or indirectly, there is an upper limit at a fixed discount rate, once the x public companies account for all gdp, it is simply divide that gdp by the discount rate. If public companies are permitted to own the shares of other public companies, then it really depends on the rules for calculating their value and avoiding the double counting of income generating assets.

Are you sure the conventional wisdom is that the value of the largest firms are expected to grow faster than GDP? Conventional among who? Sophisticated investors only need to expect their value plus reinvesting their dividends to grow faster than GDP to expect their portfolios to grow faster than GDP. Since the dividend yield has often exceeded gdp growth, sometimes the companies don’t need to grow in value at all for investors to succeed at growing their portfolios faster than gdp.

• Thank you, your final point is particularly interesting. I will take a look and maybe ask another question if I can't figure out the answer easily. – Dan Nov 28 '19 at 11:12