Investing, for real

The previous nine chapters were about the machine: how orders meet, who sits on the other side, how prices move, and why a trading bot that looks brilliant usually is not. This chapter turns all of that into what it means for your own money. It is short on secrets, because there are not any. The things that reliably help are dull, and the things that reliably hurt are expensive and exciting.

Compounding does the heavy lifting

When an investment earns a return and you leave that return invested, next year's return is earned on a bigger pile. That is compounding, and over long periods it dominates everything else. At 7% a year, money roughly doubles every ten years. Over 40 years it doubles about four times: one dollar becomes about fifteen.

US stocks have returned roughly 10% a year over the last century before inflation, and roughly 7% after it, with very large swings along the way. Nobody can promise the next century will look like the last one, and other countries' markets have had worse stretches. But the logic from Chapter 1 still holds: businesses as a whole earn profits and grow, and owners of the whole market collect that, minus costs.

Compounding works on costs too, and in the same direction. That is the next instrument.

A fee of 1% sounds small because it is quoted per year against your whole balance. But it comes out of your balance every year, including out of all the returns you would have compounded on top. Over a working life, a 1% fee commonly takes a quarter or more of what you would otherwise have had. A broad index fund from a large provider can cost a few hundredths of a percent a year.

The same arithmetic applies to every cost from earlier chapters: the spread you cross on each trade, the price impact of size, taxes on short-term gains, and advisory fees. A strategy that trades often has to beat the market by more than all of those combined, every year, just to break even with a cheap fund that does nothing.

Why professionals mostly lose to the index

An index fund holds the whole market and so earns the market's return minus a tiny fee. Every active fund manager is trying to beat that by picking better stocks. But the active managers, taken together, largely are the market. Their average result, before costs, is the market's return; after their higher fees and trading costs, it is below it. That is the zero-sum logic from Chapter 1 applied to professionals.

The data agrees. S&P Dow Jones Indices publishes a regular scorecard (called SPIVA) comparing active funds with their benchmark indexes. Over periods of 15 years or more, the large majority of actively managed US stock funds, typically around nine in ten, have trailed their index. And the few that win in one period are rarely the ones that win in the next, which is what you would expect if much of the winning is luck.

Diversification: the only free lunch

A single company can fail. Even a good one can spend a decade going nowhere. When you own hundreds of companies, the disasters in some are offset by the successes in others, and what remains is mostly the return of the market as a whole. You keep the average return and shed a large part of the risk. Economists call this the closest thing in finance to a free lunch.

It matters more than it sounds, because the returns of individual stocks are extremely lopsided. A 2018 study by Hendrik Bessembinder of every US common stock since 1926 found that most of them did worse over their lifetimes than one-month Treasury bills, and that a small minority of companies accounted for essentially all of the stock market's gains above that. If you own a few stocks, the odds are that you miss the handful that matter. If you own all of them, you cannot miss.

The middle line barely changes as you add companies. What changes is the spread around it. With one stock, the bad outcomes include losing most of your money and the good ones include spectacular gains, which is why stock tips are so seductive: the stories you hear are about the right tail. With a hundred, the range tightens around the market's return. The median one-stock outcome is also lower than the median for a basket, because a single volatile stock that falls 50% has to double just to get back to where it started.

Time is what makes stocks safe enough

Over a single year, stocks fall often. In the US, the broad market has declined in roughly one year out of four, and severe crashes of a third or more happen a few times a generation. Over longer periods, the good years have historically outweighed the bad ones by more, and the chance of ending below where you started has shrunk. It has not vanished: US stocks ended the ten years from 2000 to 2009 slightly below where they started, even counting dividends, and Japan's market took more than three decades to regain its 1989 peak.

This is the reason the money you will need in the next few years does not belong in stocks. A house deposit due next summer, or an emergency fund, should sit somewhere boring (a savings account, short-term government bonds or a money market fund), because over a single year there is roughly a one-in-four to one-in-three chance of a losing year, and it could arrive at exactly the wrong time. Money you will not touch for decades is where the stock market's return earns its keep.

The enemy is usually you

The biggest cost for many investors is not a fee. It is buying after prices have risen, because that is when it feels safe, and selling after they have fallen, because that is when it feels dangerous. Studies of fund flows regularly find that the average investor in a fund earns less than the fund itself did, because money arrives late and leaves early.

Crashes are the test. In 2008 and early 2009 US stocks fell by about half; in February and March 2020 they fell by about a third in a month. Each time, people who sold near the bottom locked in the loss, and people who held, or kept buying on schedule, recovered within about five years after 2008, and within months after 2020. A plan you have written down in advance, and an automatic monthly investment, are the two most effective defences against your own instincts.

Lump sum or a bit at a time

If you have a large sum to invest, putting it all in at once has historically beaten spreading it over a year or so roughly two-thirds of the time, simply because markets go up more often than down, so waiting usually means buying at higher prices. Vanguard has published research showing this across several countries. Spreading it out (called dollar-cost averaging) reduces the regret of investing everything the day before a crash. If that regret would make you abandon the plan entirely, the slower route can still be the better choice for you. For money you invest out of each paycheck, the question does not arise: you are already investing as soon as you can.

Taxes and the accounts that shelter them

In the US, profits from selling investments are taxed as capital gains. Gains on things you held for more than a year are taxed at lower long-term rates; gains on anything held a year or less are taxed like ordinary income. Frequent trading therefore pays a higher tax rate as well as higher costs. Dividends are taxed too.

Tax-advantaged accounts change that arithmetic dramatically. A workplace 401(k) and an individual IRA let your investments grow without yearly tax; in the traditional versions you get a tax deduction now and pay income tax when you withdraw in retirement, and in the Roth versions you pay tax now and qualified withdrawals are tax-free. Many employers match part of your 401(k) contributions, which is an immediate return no investment can compete with. Contribution limits and rules change most years, so check the current figures from the IRS or your plan before you rely on a number. Other countries have their own equivalents, such as ISAs and SIPPs in the UK.

What to actually do

This is not personal advice, and your situation (debts, income, age, other obligations) matters more than anything on this page. But the evidence points in a consistent direction for most people investing for the long term:

1. Keep money you need in the next few years, plus an emergency fund, out of the stock market.
2. Pay off high-interest debt first. A credit card at 25% is a guaranteed loss larger than any market return.
3. Use tax-advantaged accounts, and take every dollar of any employer match.
4. Hold broad, low-cost index funds rather than a handful of stocks. Mix in bonds as you near the date you need the money.
5. Invest automatically, on a schedule, and do not sell because prices fell.
6. If you want to trade or pick stocks for the interest of it, do it with a small amount you can afford to lose, and measure it honestly against what an index fund would have done.

If you want to go back through any part of the machine, the contents lists every chapter, and the bot lab is still open.