Should We Dollar-Cost Average?

During discussions with new clients, one question which almost always comes up is "should we put the money in all at once, or spread it out over time?"

If the question is instead "is it best to put it all in at once?", the answer would be yes. Strictly speaking, you can expect better returns by investing a lump sum.

But "should we" requires more thought, because each investor has their own unique circumstances, investment experience and attitude to risk.

Drip-feeding into an investment is known as "dollar-cost averaging", as the average cost of the shares you buy decreases over time. Going all in at the start is known as "lump-sum investing".

Why are lump-sums technically better?

A very common phrase thrown about by advisers is "it is time in the market, not timing the market". Two things to remember about share markets:

  1. They go up more often than they go down;

  2. Nobody knows beforehand what they will do today, next week or next month.

With this information, we should assume everyday will bring positive returns. Therefore, by investing a lump-sum as soon as possible you maximise the time invested and get the highest expected return.

Why this may not be best for you

By investing all at once, you should expect a higher return on average. However, there is the possibility markets go down right after you invest.

Spreading your investment over time lowers the average return you should expect, but reduces the range of possible outcomes. You will be reducing some downside risk while missing some of the potential (and more probable) upside.

Foregoing higher expected returns for less uncertainty is a favourable exchange for many investors. Which decision is right for you will depend on your attitude to risk.

Psychological factors

Mary Holm has covered dollar-cost averaging frequently, usually in response to investors sitting on the sidelines. In an article from August 2018 she described two important psychological factors keeping many investors in cash.

 
The first is that if someone is going to invest a large sum all at once, they will probably worry that the market will fall soon afterwards. That fear can keep people on the sidelines — in a bank account — for months, whereas if they had drip-fed the money they would probably get on with it.

The second psychological point is that lots of research shows that most people want to avoid investment losses more than they want to make gains. Let’s say there’s a choice between:

An investment that will do well most of the time, but there’s a fairly big chance it will do badly.
An investment that will bring in middling returns over all, and usually do somewhat worse than the first option.
Many people will choose the second one. To see how people dislike volatility and uncertainty, just look at all the money New Zealanders hold in bank term deposits.
— Mary Holm, NZ Herald
 

If you find yourself more concerned with potential losses than missing out on gains, dollar-cost averaging is probably better for you. This can also add discipline to how you enter the market, as long as you stick with your plan whether markets go up or down.

The difference may be small

A study by Vanguard found lump-sum investing beats dollar-cost averaging about 60% of the time, if you were buying in over a 6-month period. This increases to 92% of the time over 36-months.

As long as you don't stretch your payments out for too long, the reduction in expected returns likely won't be too large. Over 10-years, that Vanguard study found lump-sum investors ahead by between 1.5% - 2.5%, depending on the country.

2% is nothing to be sneezed at, especially with a large investment. Over ten years however, this trade-off isn't very large if it gives you a more comfortable investment experience.

Time-frames for dollar-cost averaging

As mentioned above, a lump-sum investment is almost always better than averaging in over three years. So, assuming we have decided to dollar-cost average, over how many periods should we split our deposits?

The more frequent the better, so monthly or even weekly deposits are favourable. This is because you will be buying in to the market at a wider range of prices, creating a narrower range of outcomes. Thankfully internet banking allows for automatic payments, making this option easy for investors.

As for the overall time-frame, this again depends on your attitude to risk and aversion to losses. The longer the time-frame, the longer much of your money earns a pittance in the bank.

We typically wouldn't recommend more than a year before you are fully invested. For an inexperienced investor, anxious about the market, taking a year or longer to get invested is still better than sitting in cash.

For most, six-months is a good time-frame over which to dollar-cost average. The compromise between expected returns and downside-risk is quite balanced over this amount of time.

Situational factors

When deciding whether to dollar-cost average or invest all at once, you should consider your financial position as well.

If you have a large amount of cash to invest, the decision is an important one. Maybe you have sold a house or received an inheritance. In this case, give thought to the pros and cons described above.

For those who are still working, the act of saving over time gives the same benefits as dollar-cost averaging. If your initial investment is relatively small compared to the amount you will invest in time, investing all at once is fine.

If you are moving between similar investments, such as changing between two share funds, doing so over time doesn't really benefit you. Both options are already exposed to similar risks, so moving bit by bit won't help.

The verdict

While investing all at once gives the highest expected return, for many investors, trickling money in is still a better idea. This can help nervous investors get into the market when they otherwise wouldn't.

The difference between the options won't make or break your long-term plans, so don't lose sleep over it. Instead, choose an option that works for you and stick with it. If you decide to go with dollar-cost averaging, choose a sensible time-frame and make deposits regularly.

Spotlight: Ken French

We are finding most commentary on markets and the economy, while understandably bleak, is not useful for long-term investors. To provide perspectives based instead around what we know about investment principles, we have decided to shine a spotlight on some noteworthy individuals in the investment space.

The first person we wanted to highlight is Professor Kenneth French, a key figure at Dimensional Fund Advisors and in the academic community. Ken French has a way of explaining investment principles in a straight-forward manner, useful for both new and experienced investors. We have collected the following material for your interest.


KenFrench.jpg

About Ken French

Kenneth R. French is the Roth Family Distinguished Professor of Finance at the Tuck School of Business at Dartmouth College. He is an expert on the behavior of security prices and investment strategies.

He and co-author Eugene F. Fama are well known for their research into the value effect and the three-factor model, including articles such as "The Cross-Section of Expected Stock Returns" and "Common Risk Factors in the Returns on Stocks and Bonds."

His recent research focuses on tests of asset pricing, the tradeoff between risk and return in domestic and international financial markets, and the relation between capital structure and firm value.

French has consulted with Dimensional since 1992 and is now the firm's head of investment policy and a member of its board of directors. He and Professor Gene Fama help develop and refine Dimensional's investment strategies. Fama and French also work with Dimensional's financial advisors and institutional clients to engineer efficient investment solutions.


The Difficulty of Timing the Market


In this short video, Ken French compares market timing to betting on sports.

He packs a great deal of insight into such a short speech, describing how it is not enough to predict the odds of each team winning, you must also know when the bookmaker has miscalculated their payouts.

This video is one of four released by Dimensional. Here are links to the other three in the series:


Fama/French Q&A Forum


Ken French and Eugene Fama have an online forum where they answer questions from investors. Although they do not frequently add new questions, this is because the answers they provide seldom change with time.

For an expert's view on a wide range of economic and investment topics, we recommend skimming through the forum. Here is a quick example of the questions they have answered:

 

Q: Many experts characterize the current environment as a "stock picker's market." Is there any evidence that stock selection is more successful under certain market conditions?

EFF/KRF: They can't be experts since, as Bill Sharpe pointed out in 1991, this is a fallacy of arithmetic. In aggregate, investors hold the market. This means that, before fees and expenses, the alpha for investors as a whole is always zero.

Our recent mutual fund paper says that, in aggregate, passive mutual funds nail their benchmarks: their aggregate alpha is zero. If this is true for passive investors more generally, it implies that the aggregate pre-cost alpha of active investors (stock pickers and other types) is also always zero–regardless of market conditions.

Active investors can only win at the expense of other active investors.

[Note: alpha is an excess return above the market, earned through successful active management]


Rational Reminder Podcast


For those interested in podcasts, Rational Reminder, a project by Canadian advisers PWL Capital, had Ken French on for their 100th episode. They discuss a wide range of topics, all of which are timestamped on the web-page.


A New Normal?

The phrase "new normal" has been thrown around a lot recently. Following the GFC, it was made in reference to expectations of subdued returns to come. Ken French, in the 2011 interview below, dispels these ideas. He suggests that times of uncertainty offer greater expected returns for investors.

With ideas of a "new normal" reemerging, Ken's insights remain invaluable to long-term investors.

Dimensional's US Track Record

Comparing Dimensional's US Strategy to the Overall US Market

The US share market is great for case studies as the available data goes all the way back to 1927. For the examples below, we will compare the S&P 500, the most popular index for the overall US share market, to a typical Dimensional US share strategy.

The Dimensional Strategy is made up of:

  • 1/3 Dimensional US Large Cap Index

  • 1/3 Dimensional US Large Cap Value Index

  • 1/6 Dimensional US Small Cap Index

  • 1/6 Dimensional US Small Cap Value Index

Through these weightings to small and value companies, Dimensional aims to outperform the market while remaining widely diversified.

The chart below shows the calendar year returns for both the Dimensional strategy and the S&P 500 from 1927 - 2019.

DimensionalvS&P.PNG

Over time, we see a lot of ups and downs. Also, it is difficult to see which of the two options has done better. To make it easier to judge the performance, we can subtract the S&P 500's return from Dimensional's return. When the difference is positive, Dimensional has outperformed. Sorting the years from best to worst gives the chart below.

DimensionalvS&PBesttoWorst.PNG

At a glance, it seems Dimensional's strategy outperformed the S&P 500 about half the time (actually 55% of the years shown).

It is important to note, outperformance in good years significantly outweighs underperformance in bad years. About 20% of years have an outperformance of more than 10%(!) while four have an underperformance of more than 10%.

If you are invested for the long-term, these good years stack up nicely. Below is the same chart, but showing decades instead of years. The decades overlap (e.g. 1990 - 2000 is included alongside 1991 - 2001).

DimensionalvS&PBesttoWorstDecade.PNG

This shows how the odds are stacked in Dimensional's favour over longer time frames.

There are some bad decades, but only about 20% of the total. We also see the same trend as before, where bad decades result in slight underperformance when compared to outperformance in good decades. Also, underperformance doesn't necessarily mean a loss in these examples, only a return lower than the market as a whole.

We have just experienced a decade where Dimensional's strategy has underperformed the index. It may be tempting to jump ship following this disappointing run.

First off, a common caveat in the investment industry is "past performance does not guarantee future returns". There is no reason to believe performance will suffer going forward because of what happened in the last 10 years. There may be reason to believe the opposite.

Let's look at the 17 periods, from the full set of 92 decades, where Dimensional underperformed. Then we can see how Dimensional's strategy performed in the following decade.

DimensionalvS&PWorstDecades.PNG

Historically, underperformance has not continued following a bad decade. In fact, previous bad decades have been followed by a period of outperformance.

For investors who stayed in their seats, the outperformance in the following decade was large enough to offset the previous ten years in all of these periods.

DimensionalvS&PWorstDecades20yrs.PNG

We will have to wait and see whether these trends continue, but history tells us to expect good times to come.