Month: November 2015

Monetary policy in SA – the dangers of bad theory put into practice

The Reserve Bank is well aware that there are no demand side upward pressures on the inflation rate. In fact the opposite is very much the case. Weak demand is clearly constraining the power that sellers have to increase prices, not only in South Africa but globally. Hence for inflation forecasters, including those at the Bank, lower rather than higher than expected inflation.

Moreover, the price increases to come from Eskom and municipalities and perhaps also the Government (higher tax rates) can be expected to not only add to the CPI but deflate household spending even further in the months to come. But regardless of even slower growth to come, the Monetary Policy Committee (MPC) decided on a 4-2 count that a further sacrifice of expected growth was called for.

To quote the concluding remarks of its statement of 19 November:

“In the absence of demand pressures, the MPC had to decide whether to act now or later. On the one hand, given the relative stability in the underlying core inflation, delaying the adjustment could give the MPC room to re-assess these unfolding developments at the next meeting, and avoid possible additional headwinds to the weak growth outlook. On the other hand, delaying the adjustment further could lead to second-round effects and require an even stronger monetary policy response in the future, with more severe consequences for short-term growth.

“Complicating the decision was the deteriorating economic growth outlook. Although the change to the growth forecast was marginal, the risks to the outlook, which were more or less balanced at the previous meeting, are now assessed to be on the downside. Against this difficult backdrop, the MPC decided to increase the repurchase rate by 25 basis points to 6,25 per cent per annum effective from 20 November 2015. Four members preferred an increase, while two members favoured an unchanged stance. “
The Reserve Bank has again been guided by a theory of dubious logic and unbacked by evidence that inflation expected in SA can drive inflation ever higher – regardless of the state of demand in an economy. Or in other words firms and trade unions with price and wage setting power, in their budgets and plans for the future, having set their new demands on consumers and employers with expected inflation in mind – will stick to them. Stick to them, that is, and then ask for still more at the next round regardless of the ability or willingness of customers or employers to meet these demands. Without support from the demand side of the economy and highly accommodative monetary policy responses to higher expected and actual inflation, ever higher prices cannot stick, and will not stick because such behaviour is simply not consistent with income maximising behaviour. It has not done so to date and will not do so as the theory of prices, properly understood, would predict.

In simple theories of inflation used by most central banks, including the US Fed, and as explained very clearly in an important speech given by Fed Chair Yellen recently, the influence of inflationary expectations on prices – that find the way into prices asked for – are combined with and excess demand or supply variable, known as the output gap. The theory is that the wider the output gap, the less inflation – for any given inflation expected. The Reserve Bank has seen fit to deny the role of the output gap and its own already higher interest rate settings in restraining inflation. It is a peculiar monetary policy and, more important, theory of economic behaviour that is being applied by the Reserve Bank.

Furthermore, the evidence is very strong that inflation expected in SA is highly stable about the 6% level and is likely to remain so – if and when inflation in SA trends lower. It can and should do so if the rand strengthens – a force largely beyond Reserve Bank powers or powers to predict. Inflation leads inflation expected in SA, as it did between 2003 and 2006 when, thanks to a strong recovery in the rand, inflation receded and inflation expected followed.

We can only hope for a further episode of rand recovery, lower rates of inflation in SA to follow and less inflation expected as measured by the gap between conventional and inflation linked bond yields. If this should happen then the theory of inflation being driven by the mere thought of more inflation will be thoroughly and most helpfully disabused – as it should be.

As it happened on Thursday 19 November, the chances of this test of the theory improved for reasons, surely completely independently of the Reserve Bank decision that was taken at about 15h30 our time. As Bloomberg shows (see figures below) almost exactly at that time the MPC announced higher short term rates for SA, long term interest rates in the US fell quite sharply and long term rates in SA followed lower. A surprising combination of higher short term rates in SA and lower long term rates was to be observed. Also to be observed was a stronger rand. Again, this was caused by forces quite beyond Reserve Bank influence. Emerging equity markets enjoyed a nice bounce higher – consistently combined with the lower interest rates in the US – and even more consistently combined with a firmer rand. Clearly the fear of a Fed rate hike has been well priced into markets: the Fed decision to raise rates in December has become much more certain fact and its impact much less disturbing.

This confirms once more that the most important influence on inflation, the behaviour of the rand, is largely beyond the influence of short term interest rates in SA. Therefore the Reserve Bank can in practice only hope to influence the level of demand in SA, which it does consistently by raising or lowering interest rates. Given weak demand and the absence of demand side pressures on prices in SA, it should be lowering, not raising its repo rate. The Reserve Bank is relying on a theory that inflation in SA could become self-fulfilling and therefore demands higher interest rates, doing our economy a grave disservice in the process.

 

Global rates: The dropping of the shoe

The reactions of the Fed. The other shoe has dropped – thankfully for those living downstairs.

The first Fed shoe dropped in May 2013 when it announced it would soon be halting, or in its own words “tapering” QE, that is the purchase by the Fed of US government bonds and mortgage backed paper in the market place in exchange for its own deposits. In other words, the Fed signaled its intention to stop creating money, as it had been doing to the tune of an extra US$85bn a month. True to its word, by year end 2014, QE was suspended.

The reactions in the financial markets to this announcement in 2013 were quite dramatic, and especially so in emerging market (EM) equity, currency and bond markets, including South Africa. The rand lost over 12% of its US dollar value within a few weeks, from about R9 to the dollar at the beginning of May 2013 to about R10 at month end while the yield on the RSA 10 year increased from 6.3% p.a to 7.08% p.a by month end. The benchmark EM equity index, the MSCI EM, lost 10% of its value between May and June 2013 while the US dollar value of the JSE All Share Index lost 6.8% in US dollars over the two months.

The second Fed shoe has now dropped, which is perhaps just as well for those who have been waiting for it to hit the floor. The second shoe comes in the form of the upward move in the Fed’s short term rates, the first such increase since the financial crisis of 2008. An increase of 25bps in US short term rates in December now seems certain, or at least the market place has reacted as if it is almost certain.

Market reaction to this news have been far more muted than the responses described as the taper tantrums of 2013. The rand has lost about three percent of its dollar exchange value since the September month end. 10 year bond yields are about 30bps higher in response to the now firm prospect of higher short rates in the US.

The shoe having dropped, is there more damage in prospect for the rand and the borrowing costs of the SA government? The answer will depend largely on ongoing investor sentiment towards emerging markets. Higher interest rates in the US and elsewhere are not welcome in hard-pressed EM economies. But confirmation that the US economy is firmly on a recovery track, is surely encouraging to all those EM businesses that trade with the US. A combination of strength in the US and less anxiety about the Chinese economy would surely be better news for EM economies and their longer term prospects that now appear so poor (as reflected in EM share markets that in US dollars are well down on their levels of September 2009, while the New York benchmark S&P 500 has been racking higher ever since).

The rand remains an EM equity currency. It continues to move in response to the US dollar value of the EM equity benchmark as we show below. South African events do not appear to affect the rand in any consistent or significant way.

The rand is little changed versus other EM currencies over recent days, though both the Turkish lira and Brazilian real have recovered some of their weakness against the rand. On a trade weighted basis, the rand, in line with other EM currency and equity markets, has weakened significantly since mid-year, though much of the damage occurred in August rather than very recently.

The rand’s daily moves can be well explained by global market developments independently of SA political or economic developments (which cannot anyway be regarded as favourable). For example, as we show below, the rand rate against the US dollar can be predicted to closely follow trends in the US dollar / Australian dollar exchange rate, coupled with a measure of SA sovereign risk. Sovereign risk is measured as the premium investors would pay to ensure against SA government default on its debt. The results of such a model are shown below. It suggests that the rand, now trading at over R14 to the US dollar, has overshot its predicted value of R13.50 or so.

We get a similar result and a similarly satisfactory model of the rand when we replace the influence of the Australian dollar with the MSCI EM and combine this with the credit default spread on US dollar-denominated SA debt. The rand appears somewhat oversold using these models.

It is also clear that as the prospect of higher US rates has become more certain, the risks associated with EM debt, as measured by the spreads over US government debt, have also increased. The spreads attached to SA debt have widened largely in line with EM spreads generally. SA specific risks do not appear to have had a significant influence on these spreads recently. Higher spreads and higher interest rates in SA appear mostly as an EM rather than SA event. The Credit Default Swap (CDS) spread between SA dollar-denominated debt and the average EM (Brady Bond) spread has not altered recently. In a relative sense, SA debt lost some rating ground versus other EM borrowers by mid-year, however. The spread in favour of SA can be seen to have narrowed.

Long term interest rates in SA have followed modestly higher rates in the US as the near certain increase in short rates was priced into the debt markets. As we have mentioned, these increases can be regarded as modest to date.

The wider EM risk spreads have not led to any exaggerated movements in the yields on rand-denominated RSA debt. We must hope that the very little inflation expected in the US helps to continue to restrain the Fed from ratcheting up short rates and that long term rates in the US remain at historically low levels for an extended period of time. We expect very dovish Fed reactions, especially given the stronger dollar that will keep down the pressure on metal and mineral prices and make deflation rather than inflation the focus of Fed concerns.

Less upward pressure from US interest rates on SA rates (short and long) will be helpful for the rand and the inflation outlook in SA. Hopefully, less pressure will restrain the Reserve Bank from even thinking about higher interest rates. Higher rates in SA would not necessarily protect the rand should the dollar get stronger with higher rates in the US.

The other shoe – in the form of market reactions to higher interest rates in the US – may well have dropped. And the reactions in the market place to date reflect a much more relaxed response to the prospect of higher rates in the US than was the case in 2013. We must hope and encourage the Reserve Bank to also keep its composure.

Secular Stagnation or normalisation of the global economy? Giving the patient time to return to health

A world of permanently low nominal and real interest rates, as well as permanently low inflation, is implied by the current values attached to the US equity market. Our own view is somewhat different – the world is moving gradually along the road to normalisation.

The balanced portfolio – how should it be weighted for the next 18 months. More or less in equities – Risk on or risk off.

The global portfolio manager has become noticeably uncertain about the global growth outlook and the outlook for US interest rates. The day-to-day volatility of share, bond and currency markets reveals this. More risk means higher required returns and thus lower valuations, and vice versa.

The twin concerns (global growth and US rates) are not independent of each other. Faster growth would normally lead to higher interest rates and higher rates might then be regarded as a welcome indicator of faster growth under way. Slower growth would ordinarily lead to lower rates. In a normal environment, good economic news could mean higher interest rates, but also increased revenues and profits for listed companies, so would therefore be well received in financial markets.

But the times are not normal. The developed economies (apart possibly from the US) appear to be suffering from growth that is too low for comfort and inflation rates that are too low for the comfort of central banks. The emerging economies, particularly the commodity producers, are suffering from slow growth as well as weak demand and lower prices for their exports – leading to pressure on their exchange rates. Devaluation of emerging market currencies brings higher rather than lower prices in its wake and possibly (and ill advisedly), higher interest rates that in turn would reduce growth rates further. The emerging market policy makers would prefer faster growth in the developed economies they supply and lower interest rates to take pressure off their currencies and inflation rates.

Ideally, lower interest rates designed to stimulate faster growth in the high income world, would be broadly welcome in the financial markets. There is however the problem that interest rates, more particularly real, after-inflation interest rates, are already at historically low levels. How much further can they be made to fall? There are, for practical and theoretical reasons, lower bounds to interest rates.

Quantitative easing (QE) may become the only tool available to central banks when fighting deflation, has become their primary objective and when interest rates are very low, perhaps even below zero. But even supplying more cash to the banking system may not work if the banks prefer to hold the extra cash, in the form of extra deposits with the central bank, rather than put them to work funding additional loans and overdrafts. QE may have saved the financial system, but the rate of growth in bank credit has remained very weak in Europe while somewhat more robust in the US.

In a world where prices are falling, it might take interest rates well below zero, that is well below the rate of deflation, to stimulate more borrowing and spending; that is to effectively reduce the real costs of borrowing and repaying loans that rise as prices fall. Deflation is helpful to lenders and harmful to borrowers. Inflation does the opposite, which is why expected inflation and / or deflation would always be reflected in the terms lenders and borrowers could agree upon.

Expected inflation brings higher interest rates and expected deflation would result in lower rates, even negative market-determined interest rates. In other words, you could be paid by lenders to issue debt (at negative rates of interest) as governments in Europe are being paid to do. The German government now to offers a positive rate of interest for Bunds that mature only after 2021. Recently the US Treasury has issued three month bills at a zero rate of interest, a record low.

The problem with negative interest rates imposed upon central banks is that negative rates of interest on bank deposits or other rewards for lending provided by financial institutions generally, would have to compete with cash in portfolios. Cash, or rather the notes issued by central banks as well as their deposits, will maintain their money value despite deflation, providing a highly competitive zero rate of interest, when other safe haven rates fall below zero. For wealth owners, holding cash rather than lending or spending it will not help an economy grow faster. Such problems for central banks are exacerbated when deflation is accompanied by a recession.

It is the problem with deflation, rather than inflation, that is occupying the central banks’ minds in the US and Europe. The target for the US Federal Reserve (Fed) is 2% inflation. Anything less than 2% would therefore call for lower interest rates for fear of what deflation could do to spending and economic growth.

The problem for the Fed and Fed watchers is that the Fed has strongly signaled that it will be increasing its key short term interest rates this year. But while such an increase might make sense for the US, given the economic recovery to date, it will not be helpful outside the US. It also makes less sense for the US if it leads to deflation, accompanied as it is likely to be by a stronger dollar and so more deflationary pressures inside and outside the US. There is moreover a more general concern that US growth may disappoint anyway and that any interest rate increase will not be called for.

These considerations, especially the explicit Fed concerns expressed about the state of the global economy, convinced the Fed to postpone any increase in short rates at the Federal Open Markets Committee meeting of 23 September. This decision at first was poorly received in the market place. Fed vacillation appeared to add something to the risk premium attached to equities and currencies. More recently, a weaker employment number for the US, that strengthened the case for a postponement of an interest rate increase, saw the risk premiums decline with a much better tone on the equity and currency markets, especially for emerging market currencies.

It seems clear that the developed equity markets would welcome a mixture of stable (or even lower than previously expected) interest rates in the US. The outlook for global economic growth has also been revised lower by the International Monetary Fund and other forecasters, including other central banks, making the case for lower, not higher, interest rates in the US.

Most important for portfolio selections, developed equity markets appear rather pessimistic about economic and earnings prospects. They appear to be already valued for very slow growth. Goldman Sachs, in a recent report on European equities, given an equity risk premium of 5% (that is expected returns from equities 5% above the risk free rate that is close to zero) infers that the Stoxx 600 index for European equities is priced for zero growth in earnings per share, compared to the long term average of 5.1% p.a.

If we apply the same 5% p.a equity risk premium to the S&P 500 Index, using the implied growth in earnings per share as the risk free rate, represented by the 10 year bond yield (currently about 2.10% p.a) plus 5%, less the S&P current earnings yield of 4.93%, we derive an implied permanent growth in earnings per share of about 2% p.a. This is well below the average annual growth rate realised since 1990. With 10 year US inflation-protected bonds currently offering a very low 0.55% p.a, and nominal 10 year US Treasury trading at 2.09% p.a, the compensation for bearing the risks of inflation, or inflation expected by the bond market, over the next 10 years is currently about 1.54% p.a. Thus the implied real growth in S&P earnings per share is less than 0.5% p.a. This confirms that US equities, like European equities, are currently priced for very slow growth.

It would appear that the market is expecting secular stagnation of the developed economies rather than any normalisation of growth rates. A world of permanently low nominal and real interest rates, as well as permanently low inflation, is implied by the current values attached to the US equity market.

Conclusion – our view is different

Our own view is somewhat different. While aware of what is a somewhat confused market place, we are still expecting a further gradual move to economic normalisation. This is a process that followed the global financial crisis of 2008. This will be reflected in a gradual increase in the willingness of households in developed economies to borrow to spend and of banks to lend to them. Household debt to income ratios are in continuous decline, as are household debt to debt service ratios. A decline in these debt ratios points to normalisation of household spending propensities. This process is essential, if aggregate demand is to grow at something like normal rates, given the importance of household spending for GDP in the developed world.

It seems to us that the global economic problem is one of too little demand rather than too little being produced or capable of being produced. The supply potential of developed economies is being continuously enhanced by innovation and improved technology. Addressing the problem of under spending, after the global financial crisis, we appreciate, has taken longer than normal and required very unconventional monetary policy. This may well have had something of a negative impact on business confidence and so slower growth in capital expenditure by firms that has held back economic growth. But if households came to spend more of their incomes and firms exercise more of their capital equipment, they would normally be inclined to add to their plant and machinery and perhaps also their labour forces. Capex, rather than buying back shares or engaging in acquisitions, would then make more economic sense.

Equity markets in the developed world appear undemandingly valued for current interest rates. Interest rate increases, we think, are unlikely to threaten these valuations. Any sense that the developed markets will not slip into recession or secular stagnation will be helpful for equity values. The emerging market economies, where GDP growth and particularly earnings growth, remain under pressure from lower commodity prices, may take longer to normalise. Their progress will depend on the same improved sentiment about global growth that would mean normalisation of developed economies rather than secular stagnation. Our recommendation therefore for the composition of balanced portfolios, those that mix equities, bonds, property and cash, is for a continued modest bias in favour of risk-on, rather than insurance assets.