Month: July 2011

Money and economic activity

In the eighties and early nineties a number of attempts were made to measure the relationship between various measures of money supply growth and the growth in GDP, GDE, Household Consumption Expenditure and Consumer Prices between 1966 and 1993. (82, 1984,1989, 1990b, 1993) This earlier work on the relationship between money economic activity and prices was concluded in 1990 with an attempt to separate monetary causes and effects. That is to estimate whether the money to expenditure and income link was stronger than the income to money link, given the accommodative nature of money supply responses. It was reported that the money to income link was stronger than the reverse income to money influence, using a vector auto-regression approach. (1990a). The purpose of this paper is to update this analysis to include the past twenty years of data to establish whether or not money still matters for the SA economy in the way it did.

The full paper is available here: Money supply and economic activity (2012)

The rand: No surprises

The recent strength of the rand should not have come as a surprise. The rand has continued to follow very closely the day to day the direction provided by emerging equity and bond markets.

A simple regression equation linking emerging equity markets (represented by the benchmark MSCI EM Index) to the rand explains over 95% of the daily value of the rand as we show below. This model predicted a value of R6.80 on Wednesday 27 July, slightly weaker than the R6.70 at which the rand traded that day.

This relationship seems obvious and persistent enough and very likely to continue: where emerging market (EM) equity markets go and where global growth and risk appetite take them, the rand will follow.

The explanation for the strength of the relationship is perhaps less than obvious. That the JSE All Share Index, especially when converted to US dollar, also follows the EM Index very closely, is part of the explanation. This connection that makes the JSE so highly representative of the average EM is by no means accidental. As we have pointed out, JSE earnings in US dollars follows average EM Index earnings as closely as does the Index. This is because the major companies listed on the JSE have a global and emerging market economy reach, rather than being dependent on the SA economy. And so capital tends to flow into and out of the JSE and the SA bond market, depending on the simultaneous direction being taken by the EM equity and bond markets generally.

It should moreover be recognised that the market in rands is a large, active and liquid global market. Each day up to US$20bn worth of rands is now being traded according the SA Reserve Bank. Much of this trade is conducted between third parties not directly engaged in SA trade or finance; they are presumably trading the rand because they can easily do so and are doing so because presumably the buying and selling of rands enables traders and investors to hedge exposure to Emerging Markets and their currencies that cannot be traded as easily.

This makes the rand much more of an emerging equity market currency and much less influenced by the direction of SA foreign trade and the implications that inflation differences, or purchasing power parity (PPP), can have for this trade. Exports are encouraged when the rand is undervalued – that is exchanged for the US dollar at a lower rate than its PPP equivalent; while imports are encouraged when the US dollar is cheaper and can be bought for less than its PPP equivalent value.

It all depends on where you start

The history of the rand relative to its PPP value, that is, its value explained by the difference between higher SA inflation and lower US inflation alone is shown below. In the decade of the 1990s the rand stayed close to PPP until 1995. Thereafter it depreciated at a much faster rate than PPP until the rand blew out in 2001. By then the rand was substantially undervalued relative to PPP. However by 2011, the rand was back to its 1990 PPP equivalent.

Thus if we begin the calculation of PPP in 2000 and carry it forward until now, the picture becomes very different. By 2011 the rand, compared with its PPP value, is substantially overvalued compared to its undervalued position in 2000 and 2001. The PPP value for the rand today (with January 2000 taken as the starting point) would be about R9 to the US dollar. But the strength of the rand over the last decade was due to the recoveries from two major shocks: the 2001 shock was almost completely domestic in origin – it was linked to the initiation of the asset swap facility. The shock in 2008 and the recovery thereafter represent the impact almost entirely of global forces (that is the impact of the global financial crisis on all emerging market currencies) including the rand and the subsequent recovery from this crisis.

As indicated, the rand cannot be well explained by PPP. Capital flows explain these differences from PPP and will continue to do so whatever the SA Reserve Bank might hope or try to do about this by interventions in the currency market. The market is just too big for the Bank to hope to muscle in one or other preferred direction.

These flows and their influence on the rand were severely restricted by exchange controls before 1995. Capital flows were captured within the financial rand pool and the financial rand exchange rate insulated the (commercial) rand. These controls were lifted for foreign investors in 1995 and gradually for SA investors ever since. The relief of the exchange control log jam after 1995 and panic demand for asset swaps by individuals in 2001 explains most of the persistent weakness of the rand between 1995 and 2001 (just as the global financial crisis explained the weakness and recovery in 2008-2010). By 2002 global forces had taken over the exchange value of the ZAR.

It should be appreciated that with the almost complete lifting of exchange controls and the trading appetite for rands abroad, the rand is no longer a one way bet. It is much more a bet on emerging markets. Those who like to believe that rand weakness (given higher SA inflation) is a fact of economic life should think again. And they should also appreciate that a stable rand helps to reduce inflation pressures in SA.

To view the graphs and tables referred to in the article, see Daily Ideas in the Daily View: Daily View 29 July

US earnings: The benefits of going global

A Wall Street Journal Report today by Kate Linebaugh and James Hagerty, Business Abroad Drives U.S. Profits, points to way foreign operations of leading US corporations have contributed to the very satisfactory second quarter earnings season now well under way in New York. One third through the earnings reporting season for the S&P 500 companies, earnings are the highest they have been in four years.

These S&P 500 earnings per share may well exceed $100 for the 2011 calendar year. The drivers for this earnings growth however, as the report points out, is not the struggling US economy, but the off shore operations of these companies.

About three quarters of the companies that have reported so far have done better than analysts expected. As the WSJ report states, “…Many of them – ranging from manufacturers Honeywell International Inc. and Caterpillar Inc. to drug maker Abbott Laboratories – raised their earnings forecasts for later in the year.

The report refers to the bellwether industrial giant GE that reported a 21% increase in earnings to $3.8bn for the second quarter. Yet US revenues in GE’s core industrial businesses shrank about 3.4% while international industrial revenue soared 23% to $13.4bn, accounting for about 59% of the company’s total industrial revenue. This translates into growth in capital expenditure and employment offshore rather than on US shores making the US economic recovery less likely to benefit from the financial strength of US corporations, many of whom have a strong global footprint.

We have been firmly of the view that the most compelling way to gain exposure to the global economy is via the companies listed on the S&P 500. The valuations of these companies appeared very undemanding of earnings growth, trading as they do well below their average price to trailing earnings multiples (which averaged as much as 21 times between 1980 and 2011). The current trailing S&P 500 earnings per share, calculated before higher second quarter earnings have been reported, is US$81.31. This puts the S&P on a trailing 16.5x earnings and a prospective forward PE of under 14 times earnings to be reported in the first quarter of 2012.

This advice has proven apposite as we show below. Since 1 January 2011, the S&P 500 has gained almost 7% (to 23 July) while the MSCI EM is flat and the JSE in US dollars is 2.5% weaker than on 1 January 2011.

However as we also show, S&P earnings year to date have significantly outpaced the share market index, adding to the case for investing in the S&P 500. We remain firmly of the view that the S&P 500 is still very undemanding of future earnings growth and even less demanding than it was. And the unsatisfactory state of the US economy can be expected to continue to keep down interest rates in the US (and so the competition for equities from fixed interest income)

To view the graphs and tables referred to in the article, see Daily Ideas in the Daily View: Daily View 25 July

Interest rates: Sombre is good

A sombre outlook for the domestic and the global economy was presented yesterday by Reserve Bank governor Gill Marcus, helping the case for leaving rates unchanged, as had been confidently expected. The case for higher rates any time soon was weakened by the arguments presented in the MPC statement and in the Q&A session that followed. The opportunity to lower rates was not considered, as we were informed, though perhaps it should have been. Cost push pressures on inflation were again emphasized in the statement – and second round effects of higher inflation expected (on inflation itself) were regarded as not in evidence. Inflationary expectations, as measured for the Bank, were reported as stable to lower.

The Reserve Bank expects the small breach of the upper 6% band of the inflation targets by the first quarter of 2012 but is of the view that this breach will be very temporary and therefore no reason in itself to raise the repo rate. Base effects are pushing inflation higher in the second half of 2011 and will reverse in 2012. The key to the inflation target remains the foreign exchange value of the rand and this is proving very helpful and we think will continue to be so – to the point that the Reserve Bank is over- rather than underestimating inflation.

Prominent reference was made in the statement and in the Q&A to the “core” rate of inflation, that is inflation excluding food and energy prices, which is of the order of just over 3%. An inflation target set in terms of core inflation rather than headline inflation is preferred by many of the leading authorities on inflation targeting and the Reserve Bank may well be moving in that direction. This represents an important and welcome departure from previous Governor Mboweni’s practice and rhetoric, who was determined to allow no such distinctions or escape clauses for meeting the inflation targets. This different mindset reinforces our argument for and prediction of low interest rates for longer. Moreover it raises the likelihood of generally more stable short term interest rates in the future which would be very helpful for SA business and its customers.

Closing the gap

At the Q&A, so called output gaps received interesting attention, that is the gap between potential GDP and actual levels of output. This gap is judged still to be open but is said to be closing with current growth rates ahead of potential growth. Potential output growth this year was indicated as only 3.5% with second half growth slowing down rather than picking up. This represents a very pessimistic view of SA’s long term growth potential and is not one consistent with much growth in employment and the government objectives for employment growth.

This Reserve Bank sense of potential growth is presumably derived from a limited estimate of SA’s ability to attract foreign capital, equivalent by definition to the sustainable size of the current account deficit – both usually measured as a share of GDP. We have argued that such pessimism may be unjustified and that growth can lead capital inflows that finance and sustain growth. In other words, grow faster to improve the returns on capital invested, and the capital from global sources will be forthcoming. There is a potential virtuous circle for SA that was in evidence between 2003 and 2007 (grow faster- attract more capital – sustain the value of the rand – holds down inflation and interest rates). But expressing faith or confidence in such possibilities of faster growth with less inflation aided by foreign capital, is not behaviour expected of inflation vigilant central bankers.

GDP growth is expected by the Reserve Bank to be about 3.7% in 2011 and 3.9% in 2012, increasing to 4.4% in 2013. When the presumed output gap is finally closed (on these assumptions) late in 2012 the case for raising rates will then be more confidently made. Pressures on global food inflation are however thought by the MPC to have peaked. The MPC outlook is for inflation at the upper band 6% rate by year end and is expected to remain above this 6% upper band for the inflation target until the second quarter of 2012 and receding thereafter.

Something in reserve …

Had we been at the Q&A session we would have asked the Governor and her bench of advisors why the Reserve Bank thinks it still useful to add to its already abundant supply of foreign exchange reserves. These are large enough for any conceivable emergency that might shock the SA balance of payments. The governor indicated her own confusion about the forces that drive the rand. Clearly Reserve Bank interventions in the currency market have had no predictable influence on the rand. The realised strength of the rand has made such intervention very expensive for the SA taxpayer on whose behalf the Treasury borrows rands at 6% to 8%, to offset the impact of dollar purchases on the money supply, for the Reserve Bank to invest in US dollars and Euros that return around 2% at best. It has been an expensive and futile exercise in trying to resist market forces.

We might suggest that the behaviour of the rand is not that mysterious and will continue to take its cue from global risk appetite, well reflected in emerging equity and bond markets. The well understood rand does not make it easier to predict. This remains the essential problem for monetary policy in SA, which is to hold inflation down not for its own sake but to encourage long term economic growth (lest we forget the purpose of inflation targeting).

To view the graphs and tables referred to in the article, see Daily Ideas in the Daily View: Daily View 22 July

Employment: The annual strike (that is the loss of jobs) season is well under way

The annual strike season is well under way. The usual well above inflation increases are being demanded accompanied by the usual marches, highly rhythmical toi-toing and some violent intimidation of workers, less inclined to put their jobs at risk. And after losses of production and presumably also of wages, management and unions settle on still significant increases above recent inflation rates.

The season might well be called the season of further loss of permanent jobs in the formal sector of the economy. Wages and benefits improve for those who keep their jobs, while management are strongly encouraged to proceed with operating strategies that rely less and less on unskilled labour and more on capital equipment employed.

The outcomes are plain to see in the ever widening gap between output growth and formal employment growth. This has become ever more conspicuous after 1995, due to more onerous regulation of the SA labour market (for management).

The labour saving logic practised by management is sensible enough – including their willingness to concede well above inflation increases. The logic driving union action is less obvious to those outside the ranks of union leaders and presumably their generally supportive rank and file who seem to appreciate a good fight with their bosses. One might be inclined to think, given the long established employment trends, that the leaders would rather wish to encourage employment (perhaps of their sons and daughters) and so union membership and the dues they collect with less militancy and less aggressive demands for more. Clearly there is something else at work that makes union militancy, rather than co-operation, the action that keeps the union leadership in their jobs. And so the history repeats itself: higher real employment benefits, fewer formal sector jobs and productivity gains to compensate for more expensive labour.

Shareholders by contrast have no reason to be immediately concerned about these trends, unless they fear, as they may well, the instability threatened by the growing divide between those in good jobs and those increasingly excluded from gaining access to them. But this is an issue that the management of any one firm cannot address. The reality is that management teams have adopted labour saving or especially unskilled labour saving policies that have proved to be consistently profitable and can be expected to continue to be profitable.

Over recent years the share of operating surpluses in the gross value added by the SA corporate sector has if anything tended to rise while that of employees (including managers) in the form of wages in cash and kind has tended to fall. In other words operating profits have been improving despite higher wages for those who hold on to their jobs.

The share of the operating surplus in the value added by non-financial corporations in SA and their gross cash savings is shown below. As may be seen it is a much improved picture, especially in the form of cash flow generated by these firms that has no doubt added to balance sheet strength and added value for shareholders.

The issue confronting the firms, the unions and SA generally, is how these cash flows and profits should best be employed – in reducing debt, paying dividends, making acquisitions or (much more helpfully) for economic growth adding to capital equipment or workers employed.

The answer for SA is obvious enough to all – more jobs. The uncomfortable truth is that management has no good reason to alter its ways. They are reacting to the fact of economic life in SA that the real cost of capital, in the form of a lower risk premium paid by SA firms, has come down materially, given a most helpful political transformation. Over the same period their real cost of hiring labour has increased materially.

It would seem obvious to all but those who find it convenient to deny the relationship between employment levels and employment benefits. That is to say. in the interest of more formal jobs, it is the unions that need to become less militant and more co-operative with management. The unions need to promote employment by encouraging the adoption of policies that would make for a more flexible labour market and a much more mobile labour force that could adapt appropriately to the state of the economy. Maybe only an economic policy Codesa will lead to this.


To view the graphs and tables referred to in the article, see Daily Ideas in the Daily View:
Daily View 21 July

Gold is for the risk averse, gold shares for the risk loving

The gold price in US dollars and in rands has moved ever higher while JSE listed gold shares have moved mostly sideways over recent years. The explanation seems obvious enough. The cost of mining gold in SA has risen every bit as rapidly as the price of gold while the volume of gold mined in SA has declined consistently with these higher costs and the lower grade of ore being crushed. In 1970, before the price of gold escaped the constraints of the gold standard that set the price at $35 or approximately R27 per troy ounce, SA mines produced over 600 metric tonnes of gold a year.

SA gold mines now produce fewer than 200 tonnes of gold a year and this output is falling. The only consolation in this sad tale of events is that presumably the US dollar and rand price of gold would be a lot lower had SA continued to produce as much as it did before.

But the hope that the gold mines will be able to take advantage of the higher gold price in the form of profits and dividends seems to rest eternal. Investors have consistently paid up more for the dividends actually paid by the gold mines. Currently the mines are selling at 122 times their trailing dividends or at a dividend yield of less than a third of one per cent. Gold mining accounting earnings follow no consistent pattern and have often been negative in recent years. For the record , the JSE Gold index reported earnings per share have been positive over the past 12 months and the share prices are on average 315 times their trailing earnings.

The hope clearly implicit in these extraordinary market ratings is that gold mining companies will some day, maybe soon, be able to get more valuable gold out of the ground at greatly improved margins. This makes gold shares the ones with the longest odds in the equity markets. This taste for risk may well continue to provide support for gold shares for some time to come. Presumably the risk lovers assign only a small proportion of their portfolios to this gamble.

Accordingly the relationship between the gold price (much higher) and gold shares (sideways) is therefore a very weak one and may remain so until the mines actually deliver much improved dividends. Since January 2008, for every one per cent daily move in the rand gold price, the JSE Gold Mining Index has moved on average by only 0.42% (the so called gold price beta). However the gold price explains only about 7% of the daily move in gold shares over this period. There is clearly much more than the influence of the gold price at work on gold shares. The gold price expressed in US dollars does only a marginally better job explaining the JSE Gold index (in rands) with a beta close to 0.75 but still with very low R squared or explanatory power of 0.16%.

Gold shares are clearly only for the risk lovers. Gold itself however would have served the risk averse very well over this period. This is because, on average, when the market was down the gold price was up and vice versa. The correlation between daily moves in the JSE All share and the rand gold price was a negative (-0.16). Negative correlations of this order of magnitude provide outstandingly good risk reducing diversification benefits for portfolio managers. The correlation between daily changes in the rand gold price and the S&P 500 in US dollars has been even more negative (-0.44). Thus for South Africans with exposure to developed equity markets, gold would have provided especially good insurance. For the offshore investor the correlation of daily changes in the dollar price of gold and the S&P 500 was close to zero (0.04) over the period indicating that gold would also have provided very good insurance for the US dollar investor.

Past performance may not be a guide to future performance. But if the past is anything to go by the case for investing in gold, especially for South Africans, has been greatly strengthened while the case for investing in gold shares remains at best unproven.

To view the graphs and tables referred to in the article, see Daily Ideas in the Daily View:
Daily View 20 July

The JSE, rand and emerging markets: Following the same path

The JSE and emerging equity markets (as measured by the MSCI EM Index) have moved sideways this year. However, in something that is consistent with the now very well established pattern, the two markets indices have moved closely together day to day and month to month. Furthermore the rand, confirming another well established relationship, has continued to move closely in line with the emerging market Index.

Our model of the rand/US dollar (which includes only the MSCI EM Index as its explanation) indicates that fair value for the rand/US dollar on 12 July (given the level of the EM Index) was R6.90 compared to a closing value that day of R6.85. The fit of this model run with daily data since 1 January 2009 is extraordinarily good.

The strength of the association between the value of the MSCI EM and the JSE is explained only in small part by the fact that JSE listed shares constitute 8% of the EM Index itself and therefore attract the interest of emerging market (EM) fund managers. The relationship is more fundamentally explained by the fact that the stream of JSE All Share Index earnings per share, in US dollars, approximates very closely those of the earnings per share generated by the EM universe itself. Thus the values attached to these expected streams of earnings, by global portfolio managers, are highly comparable.

Investec Securities has created its own large market cap EM Index of about 190 individual EM companies and aggregated their earnings per share to derive both an Index called the IBICEMI and its associated Index earnings per share. As may be seen below, the IBICEMI tracks the benchmark MSCI EM Index very closely and may be considered fully representative of benchmark earnings per share.

It may also be seen how closely JSE earnings per share and EM earnings per share have been related over the years. Both earnings series demonstrated extraordinarily rapid growth in earnings measured in US dollar through the boom years of 2002-2007 before the onset of the global financial crisis in 2008. Moreover EM and JSE earnings survived the crisis in much better shape than the S&P 500 and earnings per share now exceed their pre-crisis peak levels in US dollars. The JSE Index in US dollars is now above its pre crisis peak values while the MSCI EM has still to reach pre crisis levels. This re-rating of the JSE relative to the EM ( values rising ahead of earnings ) may be seen below with the JSE trading at a 14.85 times reported earnings and the IBICEMI EM Index trading at a lower 11.21 times trailing earnings.

This rerating of the JSE may be explained by the respective earnings cycles. JSE earnings per share, while following a similar path, have grown significantly faster over the past 12 months than EM earnings represented by their large caps. JSE earnings per share measured in US dollars grew by about 46% in the 12 months to June 2011 while EM earnings growth was a less robust 26% over the same period. Clearly the relatively greater dependence of the JSE on Resource companies and their earnings has served the JSE very well as metal and commodity prices moved ahead of their year ago levels.

It can be confidently predicted that the values attached to Emerging Markets and the JSE will continue to be strongly influenced by earnings reported and expected. It can also be predicted, though perhaps somewhat less confidently, that the foreign exchange value of the rand will continue to be determined in large measure by the state of emerging equity markets. The JSE, the EM equity Indexes and the rand can be confidently predicted to remain highly responsive to the outlook for the global economy, to which emerging market economies contribute most of the growth.

To view the graphs and tables referred to in the article, see Daily Ideas in the Daily View: Daily View 15 July 2011: The JSE, rand and emerging markets: Following the same path

The Hard Number Index: Satisfactory but not improving

Economic activity in SA expanded in June, but according to our Hard Number Index of Economic Activity (HNI) the pace of growth may well have slowed marginally rather than picked up momentum.

Our HNI attaches equal weights to two very up to date hard numbers, namely new vehicle sales for June, as released by NAAMSA, and the size of the note issue as at the end of June. The HNI may be compared to the Coinciding Indicator of the SA Business Cycle calculated by the SA Reserve Bank. This indicator, with a very similar lower turning point for the current cycle, is however only updated to March 2011, leaving it well behind current economic events.

Vehicle sales began a very strong recovery in late 2009. Sales of all new vehicles were particularly strong in March 2011. Actual sales that month were 53 478 units, which, since March is usually a very good month for vehicle sales, translated into a seasonally adjusted 50 101 units. Since March 2011 vehicle sales have fallen back significantly from these levels, though sales in June were modestly up on those of May. On a seasonally adjusted basis sales had fallen from the over 50 000 units sold in March to 43 108 units sold in May and recovered to 44 359 units sold in June 2011. The vehicle sales growth cycle appears to have declined significantly with the current annual growth trend around the 10% annual rate, perhaps to recede further.

We have mentioned before that the Combined Motor Holdings (CMH) share price has consistently provided a very good, even leading, indicator of the state of the new vehicle market. This relationship appears to be holding up with the CMH share price having peaked late last year, consistent with the peak in the new vehicle cycle.

While the news about vehicle sales may be regarded as somewhat less encouraging about the current state of the SA economy, the demand for and supply of notes in June is somewhat more encouraging. The Reserve Bank, when it issues notes, satisfies the extra demands of the public for notes, presumably to spend, and from the banks for cash reserves, presumably so that they are able to lend more. Banks in SA do not hold excess cash reserves of any magnitude and so the supply of notes, adjusted for cash reserve requirements, is equivalent to the money base of the system, adjusted for cash reserve requirements, or what is also described as high powered money. This makes the note issue a reliable coinciding indicator of economic activity, with the great advantage of being a highly up to date indicator.

The growth in supply of notes to the economy slowed down consistently between early 2009 and early 2011. This growth cycle appears to have picked up momentum recently. The slower growth in the supply of notes, until recently, was however offset by lower inflation, providing scope for acceleration in the growth in the real supply of notes. This growth was necessary to sustain the economic recovery under way. Now a mixture of slightly higher inflation and slightly faster growth in the note issue has helped stabilise the real money base cycle at about a four per cent rate.

If the economy is to sustain a growth rate that is still below its potential or sustainable rate, a further increase in the rate of growth in the demands for cash, well ahead of inflation is called for. No help in this regard can be expected from lower interest rates. SA does not (alas) engage in money supply targeting or quantitative easing. However it may be hoped that any increase in short term interest rates will be postponed until the demands for and supply of bank credit and the demands for the banks and the public for more cash indicate a clear case for tightening. The case for tightening based on the most recent money supply and credit numbers remains, a very week one. Faster growth in the supply of narrow money, broad money and bank credit deserves encouragement.


To view the graphs and tables referred to in the article, see Daily Ideas in the Daily View:Daily View 13 July