How Can You Forecast Performance?

Posted by Simon Moore on August 1

When you make an investment, one of your first questions is likely how will it perform?

Of course, no one can be sure how any investment will do. However, there are several things than can be helpful when thinking about investment performance.

Remember that forecasting short-term performance of under a few years is extremely challenging as the swings in the market can be impossible to predict, but over periods of multiple years, forecasts may be useful in giving a range of how a particular portfolio may do. Below are some of the things to consider when it comes to performance.

Examine Long-Term History

Fundamentally, it can be helpful to look at how the investment has done over the long-term, over decades. Short-term performance over a year or few years, isn't that helpful, for example, perhaps the investment is in a boom or a bit of a slump and will bounce back.

However, if you look over the long-run, you can get a better assessment of how the investment might perform. This is because looking over periods of decades, all the noise of particular events can even out and you may get closer to underlying performance regardless or particular good or bad events.

Now, of course, long-term historical assessments are more likely to hold up better for long-term forecasting. It's unlikely that looking back over decades will tell you how an investment will do next month, but it may offer a reasonable guide to the next few years.

This is a little bit like not judging the football team Manchester United on how they play in a single game, but looking at how they perform over a couple of recent seasons.

Consider of Ranges Of Outcomes

Risk is basically the range of outcomes that your investment could experience. Could it rise or fall by a lot? Or will the range of outcomes be narrower?

Looking back over decades it is possible to get a sense for what good and bad periods in the markets look like. This is important, because financial markets can perform differently when the economy is doing well or poorly, so making sure you are looking at different states of the economy means that you are more likely to understand both the good and bad extremes.

This means that, rather than say you expect a particular investment to deliver, for example, 5% over the next 10 years, you look at historical ranges and say that most of the time you should see returns in a range of between 2%-8%. The key here is to look at a range of outcomes rather than focus on a single figure.

This is a bit like when you're on a journey and you don't say you'll arrive at 6.13pm, you say you'll  arrive between 6pm and 6.30pm because you don't know what traffic on the motorway will be like.

Look At Relationships Between Investments

The other thing that's helpful to examine is how the investment has done relative to others over time. For example, when shares have done badly historically, bonds have often done a little better, most of the time. That relationship isn't rock solid, but is one reason combining shares and bonds can be helpful. This can be helpful in getting an idea of how different assets will move against each other in a portfolio.

Considering Value

Another angle that can be helpful in forecasting is valuation. This can be especially helpful for a bond. For example, if a high-quality government bond yields 3% and has 10 years until it matures, then it's a fair bet that the bond will yield 3% if held for the next 10 years and nothing unexpected occurs. Obviously, then if bonds yield 1% then your investment forecast becomes a little lower and if the yield is 5% then you can be a little more optimistic for future returns.

For shares, valuation is less clear cut, but still may be valuable. Unlike with bonds, shares can swing a lot in terms of profits year to year. However, looking at the P/E ratio (or earnings yield) on a share can have some value.

For example, if a share offers an earnings yield of 10% then your future returns may be better than if your earnings yield is 3%. Yet, remember that earnings can change a lot in a short period of time, so being too reliant on earnings yield can be dangerous, sometimes averaging earnings over several years can work better. Even then, predicting returns for bonds can be a little easier than for shares.

Putting It All Together

There is no magic formula in forecasting. Trying to forecast returns over very short periods, such as less than a year or two, is almost impossible. However, over multiple years, forecasts can give an indication of a range of potential outcomes.

The best forecasts consider not just the return you might achieve but also the range of good and bad outcomes. They combine historical results looked at over long periods, with current assessment of valuations and how different investments interact with each other in different parts of the economic cycle. Even then, these are only some of the issues involved in forecasting and you should take any forecast you see with a large pinch of salt, and make sure you look at the ranges of good and bad rather than just the most likely outcome.

Nonetheless, looking at how your investment might perform and your own expectations can be helpful.

Written by Simon Moore

He was previously CIO of FutureAdvisor, a US digital advisor. His most recent book Digital Wealth, explains automated investing. He studied economics at Oxford, and completed his MBA at the Kellogg School of Management.

Ready to see how your money could grow?

a month

Build a free example investment portfolio with no commitment.