Chapter 12

Investment Historic Performance

To predict the future understand the past

‘Those who cannot remember the past are condemned to repeat it.’

George Santayana

This chapter will review assets’ historic performance. Formulating an investment strategy requires investing in assets based on expectations on how they are likely to perform in the future. To predict future performance, first understand the past.

The best guide to the future is history. It provides probabilities. If an investment behaved in a certain way 80% of the time, it is fair to assume it is likely to continue doing so. The challenge is using the right past since investments behave differently during different pasts.

The objectives of this chapter are exploring and demonstrating assets’ characteristics. So far we have discussed how they should behave. Now we will see how they behave in practice.

We are using a relatively short 17-year recent history to simply illustrate assets’ behaviour. This is hardly sufficient to draw any conclusions, which require a much longer track record.

Traditional assets

Figure 12.1 shows the cumulative performance of three traditional asset classes: UK equities (FTSE 100 Index), gilts (iBoxx £ Gilts Index) and cash (UK Cash LIBOR TR 1 Month Index).¹ Inflation rate (UK RPI Index) is also included.²

Figure 12.1 Historic performance of UK equities, gilts, cash and inflation

Source: Bloomberg. FTSE 100 Index, iBoxx £ Gilts Index, UK Cash LIBOR TR 1 Month Index, UK RPI Index. January 1998 to December 2015. Based on monthly total returns, measured in £

Performance is based on monthly total returns of indices representing the assets. Each asset is assigned the value of £1 at the end of December 1997. The figure shows how £1 would have appreciated or depreciated over time.

UK equities have experienced periods of impressive gains, such as up to 2000, from 2002 to 2008 and from 2009 to 2015. They have also experienced impressive declines, such as from 2000 to 2002 and in 2008.

Relative to lines representing performance of gilts and cash, the line representing UK equities is bumpier. This demonstrates the higher volatility of equities compared to conservative assets.

The large declines illustrate equities’ downside risk. During the 2000 high-tech bubble burst, the FTSE 100 Index fell over 42% between January 2000 and March 2003. In the 2008 crisis the index lost about 40% between July 2007 and February 2009. These drawdowns measure the fall from peak to trough.³

Such 40% drawdowns just before retirement can be disastrous without time to remain invested. If you have time, after every crash the market eventually recovers. It can take a long time, but so far in history the equity market has recovered from every drop.

The index’s worst month was the infamous September 2008 when it lost over 12% as Lehman Brothers collapsed. Investors thought it was the end of the world. Well, the world did not end.

Equity returns depend on the time period. Over the entire period, equities outperformed cash, but lagged gilts. However, during some periods, equities underperformed both gilts and cash. Yet in other times equities outperformed both gilts and cash, such as the last three years. This is risk – you can make more money investing in risk assets, but you can also lose. Timing is critical.

Gilts have experienced a secular rally over the entire period. This is not surprising. As Figure 12.2 shows, 10-year gilt yield at the beginning of 1998 was above 6%, going all the way down to as low as 1.3% on January 2015.

Gilts’ strong performance was due to a combination of carry (yield) and capital gains. When yield was at a level of about 5%, this was a big part of gilts’ total returns. Then, capital gains kicked in because of a series of reductions of the base rate by the Bank of England (BOE). In the 1980s inflation fell; after the 2000 crisis the BOE cut rates to reflate the economy; and after the 2008 crisis the BOE further cut rates, as well as engaged in quantitative easing (QE).

Figure 12.2 10-year gilt yield and gilt performance

Source: Bloomberg. iBoxx £ Gilts Index, UK Generic Govt 10Y Yield. January 1998 to December 2015. Based on monthly total returns, measured in £

According to the BOE, QE is an unconventional form of monetary policy where a central bank creates new money electronically to buy financial assets, like government bonds. This process aims to directly increase private sector spending in the economy and return inflation to target.

In the future, if the UK economy expands, rates are expected to rise. If the UK economy remains sluggish – in a Japan-like scenario – rates might stay low. In any case, gilts are not projected to perform as they did in the last two decades.

Notice the line representing gilts’ performance is smoother than that of equities, but bouncier than that of cash. Gilts are more volatile than cash, but not as risky and volatile as are equities.

Finally, the line representing cash is smooth. Cash is hardly volatile. It has been losing ground relative to inflation since 2008. When the BOE aggressively cut its base rate following the 2008 crisis, cash started lagging inflation.

Figure 12.3 shows the historic base rate and cash performance. The base rate fell off a cliff from a level of 5% to 0.5% in 2008. Notice how the cash performance line’s gradient flattens after 2008.

Cash performance over the entire period is not bad. However, with the base rate’s current low level, cash is anticipated to deliver low returns until the BOE lifts the base rate. In a low-yield environment, cash is not king.

These magnificent collapses in the base rate and gilt yield explain why deposit, mortgage and annuity rates fell so sharply during recent years. The big questions are when they will rise and which new levels they will reach – what is a ‘normal’ level? The impact of ‘normalisation’ of short- and long-term rates on pensions, savings and investing will be remarkable.

Table 12.1 shows the annualised return, annualised volatility, Sharpe ratio and maximum drawdown of each asset between January 1998 and December 2015. The table confirms what Figure 12.1 is showing. Gilts delivered better risk-adjusted performance, with higher return and lower risk compared to UK equities over this specific time period.

Table 12.2 shows the same statistics but for the last three years. During different times, assets exhibit different results. Returns and volatilities are non-stationary, they change. UK equities dominated gilts during this particular time.

Figure 12.3 The base rate and cash performance

Source: Bloomberg. UK Cash LIBOR TR 1 Month Index, UK Bank of England Official Bank Rate. January 1998 to December 2015. Based on monthly total returns, measured in £

Table 12.1 Annualised return, volatility, Sharpe ratio and max drawdown

Source: Bloomberg. FTSE 100 Index, iBoxx £ Gilts Index, UK Cash LIBOR TR 1 Month Index, UK RPI Index. January 1998 to December 2015. Based on monthly total returns, measured in £

Table 12.2 Annualised return, volatility, Sharpe ratio and max drawdown

Source: Bloomberg. FTSE 100 Index, iBoxx £ Gilts Index, UK Cash LIBOR TR 1 Month Index, UK RPI Index. January 2013 to December 2015. Based on monthly total returns, measured in £

Table 12.3 Correlation matrix

Source: Bloomberg. FTSE 100 Index, iBoxx £ Gilts Index, UK Cash LIBOR TR 1 Month Index, UK RPI Index. January 1998 to June 2015. Based on monthly total returns, measured in £

Correlations

Correlation between each pair of assets is a statistical measure of how they linearly move in relation to each other.⁴ It ranges between −1 and +1. Perfect negative correlation of −1 means every time one asset moves in one direction the second asset moves in the opposite direction. Perfect positive correlation of +1 means every time one asset moves in one direction, the second asset moves in the same direction. A correlation of zero means no correlation between assets.⁵

Correlation does not mean causality; it does not say whether one asset’s movement affects that of the other. It only measures the tendency to move in the same or opposite direction on average.

Table 12.3 is a correlation matrix. The results match the theory. Equities and gilts have a negative correlation as expected since same factors have an opposite impact on their performance.

Gilts and inflation have a negative correlation as inflation negatively affects gilts. Equities and inflation have a slightly positive correlation, as performance of equities should keep up with inflation over the long term.

Additional assets

Figure 12.4 shows the performance of four asset classes: global equities (MSCI World Index), UK IG credit (iBoxx £ Non-Gilts Overall Index), global high yield (BofA Merrill Lynch Global High Yield Index) and UK property (UK IPD TR All Property Index).

High yield and global equities can be highly correlated. High yield exhibits material drawdowns, such as that in 2008. Investors who did not panic and sell at the bottom, benefited from a strong recovery.

Figure 12.4 Performance of equities, high yield, UK IG credit and property

Source: Bloomberg. MSCI World Index, iBoxx £ Non-Gilts Overall Index, BofA Merrill Lynch Global High Yield Index, UK IPD TR All Property Index. January 1998 to December 2015. Based on monthly total returns, measured in £, except for high yield measured in $

IG credit is much less risky than is high yield. However, it has some correlation with equities as similar factors affecting the two asset classes.

Commercial property’s performance seems smooth. However, because property is illiquid and properties are not traded on an exchange, the index returns are often based on appraisals (valuations). Appraisals are sticky since frequently they are based on last available price, not on current prices, which are not always available due to infrequent transactions.

The index returns are serially correlated (autocorrelated); meaning this month’s return influences next month’s return. The index’s true volatility is understated because of that.

As always, be critical about what is presented, understanding the logic behind it.

Finally, Table 12.4 shows the return and risk statistics of the four asset classes, whilst Table 12.5 shows their correlation matrix.

Table 12.4 Annualised return, volatility, Sharpe ratio and max drawdowns

Source: Bloomberg. MSCI World Index, iBoxx £ Non-Gilts Overall Index, BofA Merrill Lynch Global High Yield Index, UK IPD TR All Property Index. January 1998 to December 2015. Based on monthly total returns, measured in £, except for high yield measured in $

Table 12.5 Correlation matrix

Source: Bloomberg. MSCI World Index, iBoxx £ Non-Gilts Overall Index, BofA Merrill Lynch Global High Yield Index, UK IPD TR All Property Index. January 1998 to December 2015. Based on monthly total returns, measured in £, except for high yield measured in $

The return and risk statistics corroborate the observations from the chart. Based on volatilities, global equities and high yield are risk assets, whilst UK IG credit and property appear to be conservative. However, UK property can exhibit material drawdowns, like that between 2007 and 2009, and its reported volatility does not necessarily reflect its true risk (for example, volatility does not reveal liquidity risk).

When comparing the volatility of global equities with that of UK equities (in the previous section), note that UK equities are somewhat less volatile. Even though global equities are more diversified, the UK equity market is relatively defensive.

The correlation matrix shows the high correlation between global equities and high yield. Whilst there are no negative correlations, all the assets have imperfect correlation, which is important for diversification.

Summary

• Equities are volatile, exhibiting large drawdowns. You do not want to lose 40% just before retiring.
• Gilts have enjoyed a remarkable bull run. However, they are not expected to repeat it in the coming years, as yields are currently low and might rise.
• Cash has low volatility. But it can have low returns, lagging inflation during some periods.
• The 10-year gilt yield has come down from a level of over 6% in 1998 to below 2% in 2015. The base rate has come down from a level of over 7% in 1998 to a level of 0.5% in 2015. This is a low-yield environment. No wonder deposit, mortgage and annuity rates have fallen.
• Relative performance of different assets depends on the specific time.
• Correlation measures the tendency of assets to move in the same or opposite directions.
• High yield is a risk asset. As you should not judge a book by its cover, do not judge an asset by its name. Even when they are called bonds, studying their behaviour reveals the characteristics of assets.
• Returns of property indices can hide their true volatility. Returns are based on appraisals, not on daily prices. Always think critically about what you are presented with. Apparent facts can hide the truth.

Notes

¹ This index shows the total return (TR) of 1-month British pound LIBOR.

² UK RPI (Retail Price Index) is available on the website of the Office for National Statistics at ons.gov.uk.

³ Maximum drawdown measures an investment’s peak-to-trough fall during a specific time period. Drawdown demonstrates downside risk. However, it is based on historic experience during the measuring period.

⁴ Covariance_{(x, y)} = Σ(x_i − μ_x)(y_i − μ_y) ÷ n where x_i and y_i are the returns of x and y at time i, μ_x and μ_y are the mean returns of x and y, and n is the number of returns. ρ_{(x, y)} = Covariance_{(x, y)} ÷ σ_xσ_y, where ρ_{(x, y)} is the correlation between x and y, σ_x and σ_y are the standard deviations of x and y.

⁵ Correlation of zero does not necessarily mean that there is no relationship between assets. It means no linear relationship. The assets can have a non-linear relationship.