4. Contributing regularly: what it really protects and what is folklore

Risk 8: Horizon, contributions and withdrawals

No piece of financial advice is repeated so often or justified so badly as putting a fixed amount in every month. Everything said in its favor is jumbled together: there is a part that is true and valuable, a part that is false and very widespread, and a confusion at the outset that is worth clearing up before arguing about anything.

First, clear up the confusion

There are two different situations that go by the same name and are not the same thing:

• The saver on a salary. His money arrives in twelve installments because that is how he gets paid. Investing it every month is not a strategy: it is the only option available. There is nothing to decide and nothing to compare.
• The one who has the money already. He receives an inheritance, sells a flat, collects a severance payment, and decides to spread that sum across twelve monthly purchases instead of investing it in one go. This one really is a decision, and it is the only one the comparison that follows applies to.

Almost all the enthusiastic defense of regular contributions is built out of arguments from the second case applied to the first, or the other way round. Separate the two and the discussion becomes manageable.

The arithmetic argument that is genuinely true

Buying a fixed number of euros buys more units when the price is low and fewer when it is high. With figures: four contributions of 1,000 euros at prices of 100, 50, 50 and 100 euros buy 10, 20, 20 and 10 units, that is, 60 units for 4,000 euros. The average price paid is 66.67 euros, whereas the average price over the period was 75.

That difference is not a salesman's trick: it is a real mathematical property. By fixing the euros and letting the units vary, the average price you pay is always less than or equal to the average of the prices. With the fund back at 100, your 60 units are worth 6,000 euros on 4,000 contributed.

The arithmetic argument that collapses on its own

Now take the same method with a different price path: 100, 150, 150 and 200. Your four contributions buy 10, 6.67, 6.67 and 5 units, 28.33 in total, which at the final price of 200 are worth 5,667 euros. Whoever invested the 4,000 in one go at 100 bought 40 units and has 8,000 euros.

The advantage in the first example did not come from the method: it came from the path. When the price falls and comes back, spreading wins; when the price rises, spreading loses, and loses heavily. Since markets rise more often than they fall, spreading has an expected cost, not a benefit: while you wait, your money is in cash, outside the asset whose risk premium you wanted to collect.

The empirical evidence points the same way. The studies comparing investing in one go against spreading over twelve months, across developed markets and long historical windows, find that investing in one go wins around two thirds of the time, by a modest but systematic margin. And there is a second awkwardness: spreading does not reduce total risk either, it only postpones it. When the twelve months are up you are as invested as the other person and exposed to exactly the same thing, only with one year less of premium collected.

The argument that really does hold it up

And even so, regular contributions are an excellent idea, for reasons that have nothing to do with statistics:

• It removes the decision about when to go in, which is the decision everybody takes worst. Without an automatic calendar, the money sits in the account waiting for a better moment that almost never comes.
• It forces you to buy when it hurts. Executing the March 2020 contribution without thinking about it is an enormous behavioral achievement, and only someone who had automated it manages it.
• It caps regret. Investing everything on January 1, 2008 and watching it fall 40% destroys more investment plans than any flawed analysis. Spreading buys peace of mind, and peace of mind has an expected price that many people are right to pay.

The honest way of putting it is this: spreading your entry is behavioral insurance with an expected premium of a couple of points. If it stops you from abandoning the plan, it is cheap. What you cannot do is sell that insurance as though it improved returns, because the evidence says the opposite.

And a warning about what it does not fix

None of this touches the problem of the previous modules. Regular contributions act on the accumulation phase, where we already know that an early fall is good news. In the withdrawal phase the mechanism works in reverse and with an aggravating factor: taking out a fixed number of euros every month sells more units when the price is low, which is exactly the same mathematical property as before working against you.

A real case: the lost decade, seen by the contributor

Go back to the US investor from 2000 to 2009. Whoever had his money invested in one go at the start of 2000 finished the decade slightly below where he started, counting dividends. Whoever contributed every month of that same period, in the same index, finished clearly above what he had put in, because most of his purchases were made below the January 2000 level, including those of 2002 and those of 2008 and 2009.

Now the counterexample, which is just as important. A Japanese investor who had contributed monthly to the Nikkei from 1990 would not have been rescued by his discipline for a very long time indeed: buying cheaply works if the price ends up rising, and there it did not rise for decades. Regular contributions improve the average entry price, they do not guarantee the result. Confusing those two things is the modern version of believing a method can substitute for a risk premium.