Interest rates and asset classes

What difference does a quarter of a percent change in interest rates make? Guy Fletcher, Head of Institutional Solutions & Research at Sanlam Investments, unpacks the knock-on effect of even a small percentage change in interest rate on your investment.

Economists, media commentators and investment professionals alike spend multiple hours observing the nuances of Central Bank policy announcements, the way that the committee members voted and what their expectations are. However, interest rate policy is acknowledged to be a fairly blunt instrument in supporting economic growth and managing inflation risk, often with (at best) an impact that takes many months to take effect.

So why should we concern ourselves?

Interest rates are considered to be important since they underpin the very foundation of asset pricing – in simple terms, the price of an asset is considered to be the present value of all the future payments that asset will make, whether in the form of dividends from equities or coupons from bonds. Higher interest rates imply lower present values and vice versa.

Two observations must be made:

1. the impact that a change in short-term interest rates has on asset price formation is thus increasingly a function of the persistence of that change and
2. the sustainable impact of these changes on real long-term interest rates is, perhaps, at odds with standard monetary theory that argues that the markets will ultimately dominate central bank interventions.

So far, so good, but who sets interest rates?

Historic monetary policy

The South African Reserve Bank currently has full operational autonomy – the Monetary Policy Committee sets the levels of interest rates (commonly known as the repo rate or the rate at which commercial banks borrow rands from the Reserve Bank) within an inflation-targeting framework. However, monetary policy has evolved over the previous decades with the Governors since 1990 as follows:

• Dr CL Stals :           8 August 1989 to 7 August 1999
• Mr TT Mboweni :   8 August 1999 to 8 November 2009
• Ms G Marcus :       9 November 2009 to 8 November 2014
• Mr L Kganyago :    9 November 2014 to date

South Africa introduced inflation targeting in February 2000 after a proposal to adopt the framework in August 1999; prior to adopting this, the Reserve Bank had followed policies including: (source: SARB)

• exchange-rate targeting
• discretionary monetary policy
• monetary-aggregate targeting
• an eclectic approach (multi-method)

“The SARB acknowledges that monetary policy cannot directly contribute to economic growth and employment creation in the long run. However, by creating a stable financial environment, monetary policy fulfils an important pre-condition for the attainment of economic development”.

This clearly has implications regarding the analysis of interest rate policy since the Reserve Bank was executing interest rate changes for different reasons and under different economic conditions (witness Governor Stals’ attempt to shore up the Rand during the 1997/8 Korean crisis).  Thus, any direct inferences must be tempered with a healthy respect for prevailing policy implications.

And, similarly, it makes you wonder what was on the Public Protector’s mind when she challenged the SARB’s mandate!!

The data

Let’s have a look at the historical evidence within the South African markets to see whether interest rate changes have a direct and sustainable impact on asset price formation. In doing so, we will break the information into two sets: pre- and post-inflation targeting, the former commencing December 1995 (after the new regime had settled down), and the latter commencing February 2000. Please note that the numbers presented below are in REAL terms.

Source: IRESS, Sanlam Investments, August 2017

* The ILB numbers pre-2000 have been estimated from post-2000 relationships for the sake of completeness

** Note that the USDZAR real return is the change in the exchange rate adjusted by the difference between the SA and US CPI rates

Our first observation is that there would definitely appear to be a difference between pre- and post-inflation targeting, despite the low reduction in overall inflation of 0.7% p.a. The most obvious other ones are

1. The significant change in the annualised real exchange rate (as a consequence of stable policy)
2. The massive jump in property returns, mostly due to rental growth and reductions in vacancies
3. The significant reduction in returns for both nominal bonds and cash – this can be directly attributable to inflation-targeting yielding a far more stable environment.
4. The significant upward move in Gold (notwithstanding its benefit from a declining rand exchange rate)

It is also notable that most of the real return numbers indicated since March 2000 are somewhat higher than those put into long term strategic asset allocation models. The justification for the lower numbers is largely based on more recent experiences, where real returns (last 7 years) are anywhere from 1% to 3% lower p.a. (cash, bonds, gold, ILB’s and equity) to 8% less p.a. (property) than indicated in the table.

This is all very interesting, but what has this got to do with the MPC announcements?

Well, let’s break down each period between the times when interest rates are increasing and the times when they are decreasing. We do this in a simple fashion. Interest rates are presumed to increase until the end of the first month, during which interest rates were cut and vice versa. In this way, there are only two regimes (Up and Down) and there is no foresight (a regime only changes after a physical announcement). Results are enlightening:

Source: IRESS, Sanlam Investments: August 2017

It is notable that CPI is far higher in regimes where interest rates are increasing as opposed to decreasing – even post 2000 (since the introduction of inflation-targeting), the difference is almost 2.5% p.a. making this a far harder threshold to overcome in a real return environment.

Secondly, the interest rate UP regime can be redefined as risk-off and characterised as a defensive environment where one should elevate the interest-bearing elements of one’s portfolio. The opposite also holds true. The interest rate DOWN environment is risk-on, where a portfolio should be biased towards growth assets. The difference for equities is marked while the change for gold is most likely the consequence of its relationship with the exchange rate.

These are not guarantees, but merely observations. One refrain that is dutifully trotted out by market practitioners is that “it is different this time”. However, it is also realistic to state that it is always different, yet the results speak for themselves.

Conclusion

The marked difference in the real performance of assets between the two interest rate regimes underpins the critical nature of the MPC’s policy on interest rates. The magnitude of the change is often less important than the direction. Portfolio managers are acutely aware of this and implement their decisions accordingly.

So where are we now? Well, on 20 July the MPC cut the repo rate somewhat unexpectedly by 25 basis points, signalling the end of a 42 month UP (risk-off) environment. If we’re now entering an interest rate (DOWN) environment, should portfolios be biased once again towards growth assets?