The most obvious characteristic of a market economy is volatility, mostly a reflection of imperfect information for most economic participants, causing most to proceed unevenly over time.

Things simply do not proceed in a straight line. Even so, the volatility in the broader economy shows patterns over time, most obviously patterns of cycles, with periods of rising activity alternated with declines, and periods of accelerating growth with decelerating growth.

We like to use the identification of cycles, and their turning points, in forecasting future economic behaviour (cycles to come). Yet there always await surprises, as no two cycles are the same in amplitude or length, no economic moment is the same, indeed at all times uniqueness rules.

For those looking for cycles in the SA economy, the past ten years have offered a few surprises, indeed to such a degree that we have difficulty in recognizing the traditional landscape.

What would be traditional?
An average growth rate of 3.5% over time (established over a 100 years). Initial cyclical recovery from resource under-utilization can be faster than average, but as resource uptake progresses and pricing behaviour becomes more aggressive (by increasing margins and real wages) things slow down. As greater demands are placed on the economy than can be easily handled, strains (shortages) arise, imports increase and the current account on the balance of payments moves (deeper) into deficit. Anti-cyclical policy action may follow, forcing a growth downward adjustment until better balance has been restored. Shock developments at this time can lead to undershoots.

So how different have the past ten years been?

We apparently experienced two expansions in which the cyclical condition was partially, if not wholly, suspended for a while and these two expansions separated by a massive shock interruption.

The first expansion erred on the fast side, exceptionally so, and the second erred on the slow side, exceptionally also. All three phenomena have their exceptional explanations, in each instance deviating from the more 'normal' range.

The first expansion started as the 1999 recession ended, and proceeded relatively slowly, though not exceptionally so, through to 2003, even if it recorded exceptional currency shock (2000-2001) and the fear of repeat (2002).

Throughout this initial expansion period, the economy did mostly not achieve full potential (resource use) according to some indicators (such as skilled labour, industrial and infrastructure and building capacity, current account, credit).

Once through this turbulent period (by which time the expansion was already nearly 50 months old), the economy entered belatedly and very suddenly (as if already past the child-bearing age, yet suddenly pulling unexpected rabbits out of many hats) into a major expansionary acceleration.

This effort was accommodated by exceptionally rich external conditions (rising commodity export prices, huge capital inflows) and a strong positive sentiment response fueling business and household spending, inviting growing credit leverage excesses.

The cyclical upper turning point dates from 2007, indicative that the pace of expansion could not be maintained. The economy was overshooting its productive capacity, with shortages and delays arising. The shock denouement followed quite quickly thereafter as infrastructure shortages (electricity) kicked in, thereafter followed by a global credit crisis also connecting locally.

Afterwards, one had to conclude that the domestic credit cycle had far exceeded normal boundaries, just as capital inflows had, with the current account accommodatingly deeper into deficit.

Leverage had been allowed beyond the accepted norm, certainly in property even if not quite like conditions in some overseas countries.

Just as the belated cyclical acceleration of 2004-2007 had been exceptional, so the shock adjustments of 2008-2009 were similarly off the chart (global talk being of the Great Recession).

With this as background, once past the lower cyclical turning point we entered the post-2008 recovery expansion. Globally, but also locally, it lived up to the billing of Reinart and Rogoff, according to whose study of financial crises this past Millennium the bigger the crisis, the slower and prolonged the recovery process.

In the case of SA there was an initial inventory-assisted revival, which looked normal, yet by March 2014 (nearly 55 months past the cyclical trough), the economy was still  heavily undershooting, with resource slack remaining in labour, industrial and physical capacity (but not electricity or export commodity rail).

In the case of residential building activity, the building trades had flattened out during the 2009 recession and thereafter stayed effectively at half cyclical peak levels of 2006-2007.

Though this was exceptional, and traceable to exceptional reasons (such as the crisis-induced change in household credit provisioning), there were a few too many surprises like that.

Mining output not deviating much from 2000 levels, except for the gold sunset condition and major iron ore expansion.

Electricity output staying unchanged near 2007 levels.

Skilled labour, manufacturing activity and capacity utilization recovering only very slowly, battling to reach pre-recession levels.

Normality, if that was what it was, could be observed in non-risk public sector manning levels and real wage gains, increases in under-regulated unsecured consumer lending and strong fiscal social spending, together feeding into strong retail and motor trade performances.

But these did not feed off "normal" private risk-taking, and as such were a poor barometer of what really was going on in the market economy.

The most historically astounding phenomenon occurred in expressed business confidence. Ever since these BER opinion surveys have been run in SA in the post-WW2 period, the cyclical behaviour of this time series was extremely pronounced and regular. In particular, after a recession (mild or otherwise), there would follow a meteoric revival (from deep negative sentiment to highly positive near the cyclical peak in a near uninterrupted advance).

Not this time. After 55 months of cyclical "expansion" this metric still lingers below the 50 (neutral) mark, in an unprecedented deviation from more normal cyclical expansions.

Together with some of the other data already mentioned, and the undershooting GDP growth (especially lately), it suggests an anything-but-normal business expansion.

The politicians like to point to exceptionally slow global revival as the main explanatory variable. Local practitioners prefer to point to poor domestic planning and infrastructure constraints, intruding government regulatory interventions, union excesses and sub-par business confidence holding back risk-taking and thereby also constraining household income growth opportunities and spending potential.

With frankly as yet no end in sight to such undershooting, as global recuperation is proceeding very slowly, and the domestic conditions described easily becoming extended until better political leadership, more restrained union behaviour, more growth-friendly policies, a reformed public sector, faster infrastructure rollout, higher business confidence and perhaps a new commodity windfall condition (shale gas fracking in the 2020s) make an appearance.

Thus the exceptional slow pace of expansion as observed since 2009 will most likely also end eventually through the means of orchestrated interventions and otherwise pure luck.

Still, we are left with three remarkable outlier events, varying from extremely fast to extremely slow expansion, interrupted by an exceptional series of domestic and external shocks.

Perhaps then still as an afterthought, if we want to continue testing the presence of stylistic cyclical behaviour, the metrics of which can be obtained from using available national accounts data going back only a few decades, I would like to make one further observation.

If the decade 2004-2014 has been exceptional cyclically as described (and not yet fully completed), a similar exercise for the period 1960-1994 will also throw up plenty exceptional cyclical fragments, both of the under- and overshooting variety and of shock-induced interruptions playing havoc with stylised cyclical expectations, especially when prematurely cutting short expansions.

In that long period SA fought a rearguard action while defending a deadend political philosophy. Something similar has also been true for the past decade and more where other nationalists pursued other policy hobby horses with similar deadening effects for the economyresulting in major and prolonged undershoot episodes, while any remarkable overshoots were generally associated with extremely fortuitous external conditions.

Thus post-1950 Prof De Kiewiet's pre-1945 dictum lived on with remarkable cyclical longevity, that modern South Africa progresses through economic windfalls and political disasters.

As the statistical economic record tends to coincide with these many repeated remarkable occurrences inclined to give rise to periods of exceptional over- and undershooting episodes in SA economic conditions, I would  be more than ordinary cautious about averaging all these observations out and deduce stylistic cyclical characteristics from them and describe these as ordinary, normal or average.

The moral of this long tale is that South Africa's modern condition is anything but normal, just as its pre-modern history had been, only decidedly different.

But, hopefully, most of us already knew that.

Cees Bruggemans

Consulting Economist

Bruggemans & Associates

Website   www.bruggemans.co.za

Email   economics@bruggemans.co.za

Twitter   @ceesbruggemans