Don't bet on one horse.
or: how spreading your bets quietly does the work
What an index actually is
An is a list. A curated set of companies that, together, are meant to represent something: a country, a sector, a global market. The S&P 500 is a list of 500 large US companies. The MSCI World tracks ~1,500 large and mid-cap stocks across 23 developed markets. You don't buy an index; you buy an that tracks it. The index itself is a thermometer; the ETF is the bottle of water with the same temperature.
500 of the largest US companies, weighted by .
→~1,500 large and mid-cap stocks across 23 developed markets: US, Europe, Japan, Australia.
→100 of the largest companies listed on the London Stock Exchange.
→225 large companies listed on the Tokyo Stock Exchange, the headline Japanese index.
→One stock, fifty, five hundred
is how much the price swings around its average. The fewer companies you own, the more your fortunes depend on each one's headlines: earnings misses, lawsuits, scandals. Add more companies, and those individual storms cancel each other out. What's left is the broad weather: how the whole market behaves. The numbers below are typical annualized volatilities for diversified equity portfolios.
A single company. A bad earnings call, a CEO mistake, a regulator decision: any one of them can cut your value in half. You own all of the upside; you also own all of the bombs.
Sector-wide trouble still shows up, but no single name can sink you. About 90% of the volatility benefit of diversification is captured by 50 well-chosen names. Most ETFs have hundreds.
What's left is essentially the market itself. The S&P 500's annualized volatility is in this range. Adding the next 4,500 stocks (a global index) trims a little more, but most of the work is already done.
The free lunch
Mathematically: the volatility of a portfolio of assets falls roughly with the square root of the number of holdings. The expected return doesn't. So you keep the upside while throwing away most of the noise. Harry Markowitz won a Nobel Prize for formalizing this in 1952. The practical lesson is much shorter: own many, not one.
Same long-run growth. Far less drama.
Diversification doesn't lower the long-run growth you can expect from owning productive companies. It lowers the year-to-year ups and downs around that growth. The expected destination stays; the road becomes far less bumpy. That's why a global stock index is the default starting point for most long-term investors.
Most of the smoothing happens fast: once you own about 30 different companies, you've already gotten ~80% of it. What's left below the curve is the market itself, a part you can't smooth out, no matter how many stocks you add.
Math behind the curve: the swings of N companies shrink roughly with √N. (Markowitz, 1952.)
The three dimensions of diversification
There are three different bets you spread when you diversify well. Most investors handle the first two by buying a broad ETF. The third (temporal) is what makes the long horizon work, covered properly in Lesson 11 on .
One country goes through a bad decade.
A global index (MSCI World, MSCI ACWI) instead of a single-country one.
One industry collapses (banks 2008, dot-coms 2001).
A broad index that owns every sector in proportion, not a thematic ETF.
You buy everything the day before a .
Buy regularly over years, not all at once. (.)
Owning everything is safer than picking the winner.
Stock-picking is a tournament most people lose. Owning the whole list, through a cheap, broad-market , sidesteps the tournament entirely. You don't need to be right about which company wins the next decade; you just need to own the basket that contains it.