Current business cycle surprises

 EmailPrint This Post Print This Post

By Cees Bruggemans, Chief Economist of FNB.

Economic activity tends to proceed in waves or cycles. No two cycles are exactly the same, but similar processes can always be identified.

Still, how different or unique is the present, and what does it promise for the years ahead?

What is currently typical (and conventionally boring) and atypically exciting (or lethal) in the South African business cycle playing out?


A business cycle can start fast or slow and end in a whimper or with abrupt suddenness, with cruising plateaus of varying lengths.

Though all these features when present will have a reason, it is not these I wish to focus on here.

Rather, I wonder about features that stand out, breaking typical patterns. And why this may be so.


Overall, the South African economy has averaged 3.5% GDP growth these past 90 years.

In the post-WW2 period, our business cycles have been mostly relatively short in duration (measured peak-to-peak or trough-to-trough).

Cycles pre-1960 averaged less than three years and during 1970-2000 averaged double that at just over five years.

Exceptions were the 1960s and 2000s decades, with 10 year long expansions, half of which at over 5% growth.

These varying lengths by themselves identify something bigger that is exceptional.

The 1960s were the tailend of a very long global high growth period that favoured South Africa, without its politics yet becoming a debilitating factor.

The 2000s saw South Africa benefiting once again from exceptional global commodity and capital flow conditions, dissolving her balance of payments constraint on domestic performance, by then again without the drawback of being a political outcast.

These two single decades between them book-ended three decades of political turmoil, external sanctions, financial overreach, loss of confidence and shortened time horizons, creating resource constraints and fanning imbalances which kept the business cycle short and average growth below par.

Also, the global business cycle was volatile during this period, with 1974 and 1980-1982 recessions and 1998 marked by the Asian Contagion crisis. An exogenous factor (severe drought) deepened our 1992/93 recession further.


Within cycles you expect every sector to experience an upswing and downswing, even if the timing can differ.

For example, residential building activity ‘leads’ non-residential building activity by 18-24 months, when cyclically accelerating (‘going up’) and decelerating (‘coming down’).

This is mainly a function of the size of the building projects in question and how long it takes to completion. A house may take a year, a large building two to three years.

The amplification of the cycle can also differ greatly. Classically, household consumption classifies goods purchases as being durable, semi-durable and non-durable, depending on length of economic life.

The more durable the goods in question, the longer the effective life but also the higher the likely price tag (a house, car, furniture, appliances), the more disturbed the replacement cycle potentially can become as consumers can decide to delay/postpone in uncertain times (and many doing so herd-like), but also bunching as they decide to end their abstinence.

Thus the amplitude of the growth cycle for durable goods is most extreme, varying potentially from +15% to -15% (or even double that in extremis), compared to half that amplitude for semi-durable goods and half that again for non-durable goods (the latter varying minimally, mostly in line with stable population and urbanization growth, except during exceptional shock-like disturbances).

Similarly, growth patterns in consumer spending tend to closely mirror income trends (the latter stabilized in modern economies by state support and other buffers).

This in contrast to fixed investment, which tends to be far more cyclical, as private businesses may have to scramble to expand capacity (on the accelerator principle) compared to cutting back heavily once a recession creates resource slack.

Public sector fixed investment should be far less cyclical, given long lead times and long state planning horizons, unless incompetence yields different outcomes.


So here we have MANY dimensions that can inject varying cyclical behaviour into an overall business cycle for a given economy, itself of course still heavily influenced by (trade and financial) cycle happenings elsewhere in the world.

Given all that, is there still more scope to identify something ‘odd’ (atypical) at any given time in any particular sector, and is there reason for the oddness to persist or should it revert back to normal (typical) in the next cycle (or so), bearing in mind no two cycles are the same?

I submit that there are currently what appear to be major atypical cyclical features. This can mean an awful lot to those so affected. And, yes, one does not expect oddness to hang around too long, though different oddities can surface in future cycles (in turn affecting new classes of producers and consumers).

Does it matter? To the planning horizons, operational demands, investment commitments, bottom lines and peace of mind of those affected, certainly.


So what’s so odd that it stands out on Google Earth?

Wanting to keep things simple, I would like to focus on opinion surveys undertaken over many decades by the BER, with RMB sponsoring the overall RMB/BER business confidence index.

It uses proven methodology (German in origin), closely tracks the economic cycle in terms of GDP growth, and gives interesting cyclical patterns, some of which have recently stood out a mile for their ‘oddity’, given the company they are keeping.

As a control group, business confidence in manufacturing, motor trade and wholesaling have given fairly typical (‘normal’) patterns over five business cycles covering forty years to date, mirroring broadly highs and lows, and the pace of ascent and descent relative to the overall business cycles.

These cyclical patterns were all present as well in the other two sectors surveyed (retailing and building contractors), except for a few ‘oddities’.

In addition, the consumer confidence survey undertaken by the BER and sponsored by FNB, also tracks the business (GDP) cycle closely, leading with one or two quarters at upper and lower turning points.

And it has had its own share of oddities which raise deeper questions about the economy.


Perhaps first the FNB/BER consumer confidence index. Two episodes in this survey’s recent history beg explanation.

Black and White consumer confidence used to closely correlate cyclically pre-1994 (the ‘control’ period).

During 1993-1999, however, a very persistent gap of as much as 40 index points (on indices that fluctuate 95% of the time between readings of -40 and +30) opened up between the two groupings for a six-year period.

For the duration, Black consumer confidence appeared about 20 points higher than a typical historic pattern would have suggested, and White confidence was at least 20 points lower than its typical historic pattern, for that stage of the cycle.

It takes no rocket science to surmise that Black South Africans were for a while exuberantly confident following the 1994 election, with its afterglow lasting a full five years. In contrast, White South Africans were probably for the same reasons considerably less confident about many things.

Perhaps more significantly, following the financial interruption and growth slowdown of 1998-1999 (marked by prime 25.5% once the Asian Contagion washed ashore here, too), both groups of consumers seemed to come to their senses, fairly quickly converging once again to proceed much more ‘normally’ and sedately for the next few years, though in some quarterly instances sizeable gaps did open up, only to erode again subsequently.

This was an ‘easy’ exception to explain.

Less easy is the behaviour of consumer confidence during 2009-2010. Following a normal peaking and falloff in consumer confidence during 2007-2008, closely mirroring the GDP business cycle behaviour, there was an abrupt change in 2009-2010.

Instead of consumer confidence falling away more decidedly in 2009, well into negative territory, mirroring the recession and previous such episodes, the overall FNB/BER consumer confidence index did not do so.

Instead, the cyclical trough effectively hugged the zero line (50/50 confidence territory), and then started to improve very early, leading the start of business recovery by 18 months (indeed even before the recession had arrived!), a most unusual occurrence.

Furthermore, by early 2010, barely one quarter into business recovery, the FNB/BER consumer confidence index was already at +15, a value usually achieved only at the peak of business cycles, and at least a few years into the cycle.

That consumer confidence wasn’t even higher, equating to peak exuberant growth cycle territory (2007) was due entirely to a healthy sense of caution reflected in the answer to the question ‘is now a good time to buy durables’. The recession, the credit crunch, events overseas, at least made a majority of consumers VERY cautious for some considerable while (this form of gloom only starting to lift from mid-2010).

But when it came to confidence readings about economic prospects and own financial prospects, the confidence readings were much higher than the overall index and equating with 2007 prosperity territory.

It was as if the shocks of 2008 (electricity, Rand, interest rates, recession) were a mere intrusion, an inconvenience that was, like a shaggy dog throwing off water, quickly left behind, resuming the very high prosperity levels of confidence achieved from 2005.

Prosperity apparently was never lost. It was merely very temporarily dented before springing back, at least in confidence terms, except for the noted caution about buying big ticket items.

This does not jell close with the unemployment losses, the credit strains, the houses and cars foreclosed, the record bank and retailer bad debt provisioning, and the general funeral atmosphere of those years.

When revisiting Black and White consumer confidence for this period, two things stand out.

White consumer confidence followed a more typical cyclical pattern fluctuating around the long-term average (near the zero line – 50/50). Black consumer confidence, however, during 2008-2010 seems to have moved higher than its long-term cyclical pattern would have predicted.

This suggests Black consumers as a group being better off and more confident about their own finances during a very difficult period, suggesting a structural shift in their financial position and sense of what was happening to the economy.

One reason for this trend is that many of the ‘emerging’ new Black middle class is being employed in the public sector and thus not exposed to job insecurity, while many Whites are today in the private sector or run their own business exposing them more to uncertainty.

The credit strains were perhaps more proportionally experienced by the old White middle class, and less severely by the new Black middle class, respectively possibly reflecting relative over- and under-indulgence in especially the credit and property spheres.

Black consumer confidence, especially in the higher income categories (over R5000 and over R10 000 monthly) may have ‘carried’ overall consumer confidence at high levels during this period. For many such consumers these levels of income must have been satisfactory, given past realities for many.

Indeed, for them the party was never really over, after the merest (if shocking) interruption of 2008.

Why doesn’t that come as a surprise? A simple look around any large town, especially large shopping malls, offers highly suggestive anecdotal evidence that this proposition may not be as farfetched as what it sounds.


When we then turn to BER business confidence data, one anomaly jumping out is that retailing TWICE showed what appeared to be an odd pattern in five cycles.

In 1994-1997, its cyclical upswing never reached typical exuberant territory at the peak, instead getting stuck just over the zero line (in positive territory).

One reason could be that at that early stage Black consumers had not yet benefited greatly from the new dispensation with White consumers still ‘overrepresented’ in the high income group.

And with White confidence deeply compromised, and their spending possibly below par, retailers in turn experienced a less than satisfactory business upswing.

Manufacturing, motor trade and wholesaling business confidence show only the faintest echo of this phenomenon, while retailing shows it starkly.

Building contractor confidence also showed a sub-par peak reading in that period, but that was merely the extension of a decade long deterioration in this industry which at the time had yet to end, White consumer confidence (or the relative lack of it immediately post-1994) yet again being the main reason.


What difference a cycle of freedom and spreading prosperity makes. For fast forward fifteen years to the recent cyclical downswing of 2008-2009, and what do we get? A VERY different confidence response from retailers as opposed to building contractors.

Retailers, like consumers, apparently didn’t experience a (deep) recession in 2009, with their business confidence abruptly collapsing from peak prosperity levels in 2008, only to get arrested near the zero line, with no deep or prolonged descent into negative territory typical of their usual business downswing and recession.

The major reason for this may be rising per capita income due to a period of high growth, employment, positive real wage increases, a lower personal tax burden (though increased other imposts) and increased social grants.

Household consumption, retail sales volumes and motor trade all suggest typical recessionary conditions, though year-on-year retail volumes contracted by a maximum 5% in 2009 compared to 10% in 1985 and 1989 and 7.5% in 1992 (making the 2009 retail recession shallower compared to earlier cyclical downswings).

Retail space did become heavily overbuild in the recent recession. Yet retailer sentiment barely flagged the slowdown, and then very fleetingly, before reviving anew, already at very high confidence levels very early in cyclical recovery.

Perhaps we are just having more happy retailers in active therapy. Then again the parallel with consumer confidence is stunning. It is just this disconnect with historic patterns (and undeniable recession, at least for some) which defies these results.

Then again I have met with at least three large retail chains whose senior staff admitted over 18 months ago “if this is a recession, please give us more of it”.

If there was a severe cyclical loss of income and buying power, retail confidence levels showed the cyclical outline, but did not come close to reflecting its usual signature. Odd, really odd, unless we all misread the nature of this recent cyclical consumption interruption.


Just to make things interesting, I suppose, building contractors swung to the other extreme. From genuine dizzily high confidence levels for very long (2003-2007 WAS extremely exceptional in this trade, admittedly after decades of drought), building contractor confidence plunged typically as the downswing hit, BUT THEN NEVER CAME UP FOR AIR WHEN THEY SHOULD HAVE.

Building activity and building contractor confidence has a very close inverse relationship with interest rates.

As interest rates peak (they did in late 2008) and start to decline, ere long building confidence should start to turn. Not this time.

A full two years into the most aggressive interest rate cutting cycle in decades, building confidence is still dropping in search of a trough.

The reason is structural. The National Credit Act was introduced in 2007, banks had the fright of their lives with Anglo-Saxon subprime mortgages and banks generally blowing up in 2007-2008 inviting a return to DEEP credit basics, and many overborrowed households probably did an inventory of their worldly possessions and came to the shocking realization they were overstocked in houses, cars, furniture, plasma screens, children, in-laws and mortgage and creditcard debt.

Between these three forces, credit growth came to a grinding halt and has taken forever to start reviving again, and then very, very slowly, this week reaching all of 5% year-on-year (with prime by now lowered to 9%, the lowest level since February 1974).

And whereas the car has a limited life, ages relatively quickly and requires maintenance while newer models keep temptingly appearing, one needs wheels as a matter of necessity for work, leisure and fun.

The car’s replacement decision, although it can be delayed, won’t be delayed indefinitely once finances improve and household anxieties diminish.

Property, and new building activity, however, encountered a steep new hurdle in credit restraint that will take time (possibly years) fully eroding away, but meanwhile distorting supply and demand, holding back recovery in confidence and activity.


All these ‘atypical’ features of recent years ultimately can be explained by events (if correctly interpreted), and should be accepted as part of an unfolding business cycle.

Yet the highlighted ones have stood out for me like a sore thumb, not least for the exceptional happiness and deepest despair I have encountered in these various sectoral spaces, their exceptional nature itself deviant from the average cyclical norms found elsewhere.

It is this what makes it atypical for me, with deep consequences, for good and bad, for those so affected. In time I expect all these departures from the typical to erode away. And yet, and yet.

Some of it (the credit constraint, income redistribution) may make this particular bunch of cyclical oddities more long-lasting atypical than I could ever dream. In which case they would become typical?

One would hope not for that could mean lopsidedness to our economic life that might neither be ‘normal’ or advisable, in the long run.

Source: Cees Bruggemans, FNB, November 30, 2010.

Did you enjoy this post? If so, click here to subscribe to updates to Investment Postcards from Cape Town by e-mail.

OverSeas Radio Network

Leave a Reply

You can use these HTML tags

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>




Top 100 Financial Blogs

Recent Posts

Charts & Indexes

Gold Price (US$)

Don Coxe’s Weekly Webcast

Podcast – Dow Jones

One minute - every hour - weekdays
(requires Windows Media Player)
newsflashr network
National Debt Clock

Calendar of Posts

Feed the Bull