Secular stagnation and the policy response: any parallels in South Africa? An address by Mr Daniel Mminele, Deputy Governor of the South African Reserve Bank, at the RMB/Morgan Stanley ‘Big Five’ Investor Conference, Cape Town
1. Introduction
Ladies and gentlemen, good morning,
It is a pleasure and an honour to address this year’s ‘Big Five’ Investor Conference, and I should start by thanking our hosts, RMB and Morgan Stanley, for inviting the South African Reserve Bank (SARB) to be part of the proceedings.
The topic which I have been asked to discuss today – that of secular stagnation and the type of policy response required in such circumstances – is highly relevant, not just for advanced economies but also for an emerging country like South Africa.
For a while now, there has been little excitement about global economic prospects, the story essentially being one of a continued subdued recovery not short of downside risks in an environment characterised by relatively high levels of uncertainty and volatility. Global economic growth has generally been soft since the Global Financial Crisis and has consistently fallen short of consensus expectations for a ‘return to trend’.
This recurrent problem has led both public- and private-sector economists to search for explanations to this apparent medium-term economic stagnation, to separate its cyclical and structural drivers, and to look for the most appropriate policy responses. As you are well aware, what exactly constitutes the ‘new normal’ or the ‘new trend’ continues to be a subject of much debate. Concerns are also being raised that too much is being expected of monetary policy in dealing with the problem. In South Africa, too, growth has slowed by more than was generally expected a few years ago, and the debate about appropriate policy responses is equally ongoing.
I shall begin my remarks today with some general observations about secular stagnation, followed by looking at the causes of slower growth in South Africa before addressing the role of monetary policy in terms of appropriate responses; I will end with a brief outlook for monetary policy in South Africa. Many of you would have read the statement released less than a week ago after the conclusion of the most recent meeting of the Monetary Policy Committee (MPC).
2. The debate about secular stagnation
Economic developments in advanced economies continue to challenge conventional economic wisdom. In the last five years, on average, economic growth in the United States (US) has posted moderate annual gains of 2,0 per cent, well short of the 3,2 per cent average posted in the 15 years prior to the 2008/09 recession. In the eurozone and Japan, economic performance has been even more disappointing.
Furthermore, while employment creation has been dynamic in the US and has improved in the eurozone (albeit to a lesser extent), productivity growth has lagged, falling well short of pre-recession norms. Increasingly, economists estimate that a large part of this productivity and economic growth slowdown is structural.
The Organization for Economic Co-operation and Development (OECD), for instance, estimates that potential GDP1 growth among its members will have slowed to 1,6 per cent this year from the average paces of 2,0 per cent in the period between 2001 and 2010 and 2,8 per cent in the period from 1991 to 2000.
While the evidence of sustainably slower growth is there for everyone to see, the drivers of that slowdown remain the object of intense debate. Some economists talk of ‘secular stagnation’, which incidentally is not a new expression, having been coined by economist Alvin Hansen in the 1930s.
Former US Treasury Secretary Larry Summers recently summarised ‘secular stagnation’ as a situation where an imbalance occurs ‘resulting from an increasing propensity to save and a decreasing propensity to invest’, the result being that ‘excessive saving acts as a drag on demand, reducing growth and inflation, and the imbalance between savings and investment pulls down real interest rates’.
This would suggest that it is a lack of demand – be it consumer or corporate investment demand – that causes the weakness in trend growth and productivity. Whether this is a temporary consequence of a ‘debt overhang’ (specifically the high degree of consumer and corporate leveraging up to the Global Financial Crisis) or a more permanent behavioural change remains open to debate.
Other economists, though, argue that the slowdown in productivity is of a more supply-driven nature. For the likes of Robert Gordon, the ‘low-hanging fruits’ of innovation have already been picked and we are now entering a period of reduced technological progress, the consequence of which will be a structurally lower pace of total factor productivity.
While such a proposition remains controversial, other economists point towards a lower-than-expected diffusion of previous technological gains – especially from firms operating at the ‘technological frontier’ to the less productive ones – or to a decreasing rate of business start-ups as explanations for the trend slowdown in productivity.
Whereas the causes of this trend growth slowdown remain open to debate, the consequences for monetary policymakers in advanced economies have been relatively straightforward. Faced with lower potential real GDP growth and with difficulties in lifting inflation rates to what were historically regarded as low targets (to around or slightly below 2 per cent), central banks have maintained looser monetary policies for longer than in previous cycles.
Equally, they have scaled down their estimate of the ‘neutral’ real interest rate. A couple of months ago, former Federal Reserve Chairman Ben Bernanke highlighted how persistent undershoots of economic growth and inflation had led the US Federal Reserve to scale down its estimate of the long-term equilibrium Fed funds rate from 4,25 per cent in 2012 to only 3,0 per cent, a level which the FOMC last week trimmed to 2,9 per cent.5 Effectively, central banks have been concerned that the monetary policies implemented in recent years – because of a lower neutral rate – may not have been as loose as required by economic circumstances.
Furthermore, they have been worried that lower equilibrium rates could limit the room for policy stimulus through conventional means in the event of a new negative shock hitting the advanced economies.
3. The causes of South Africa’s slower growth
To what extent do these advanced economy developments apply to South Africa? Looking at the economic performance of the past few years, there is no denying that domestic real GDP growth has slowed and, in a similar fashion to developed economies, it has also consistently fallen short of forecasts for even a moderate recovery in the year or two ahead.
From an average of 4,8 per cent in the five years prior to the 2008/09 recession, GDP growth slowed to an average of only 2,1 per cent in the period between 2010 and 2015 – and it is not expected to exceed 0,4 per cent in 2016, a rate which entails falling per capita income. A significant part of this slowdown is estimated to be structural. Consequently, the SARB has scaled down its estimate of potential real GDP growth to 1,4 per cent this year from more than 2,0 per cent five years ago and as high as 4,0 per cent prior to the Global Financial Crisis. Average worker productivity has echoed some of the trends of advanced economies, slowing from an average of 2,6 per cent in the 2000s to as low as a negative 1,1 per cent year on year in the first quarter of this year.
However, it would be misleading to attribute this slowdown in South African trend growth and productivity to the same causes (identified or suspected) as in advanced economies. True: the slowdown in global growth has had direct implications for South Africa, especially as growth in advanced economies has become less import-intensive, limiting the room for relatively small, open emerging market economies to engineer export-led recoveries in the aftermath of the Global Financial Crisis.
As of the second quarter of 2016, South Africa’s export volumes were only 2,6 per cent above their 2008 peak. Equally, the slump in commodity prices that commenced in 2011 has weighed on the value of South African shipments abroad and on corporate profits, in turn undermining employment and capital spending, especially in the mining sector.
Nevertheless, some of the explanations most commonly put forward to explain weak growth in advanced economies do not seem to apply to South Africa. For example, on the demand side, there is no evidence that the 2008/09 recession has ushered in a higher propensity among households to save a larger share of their income.
Over the past five years, the household saving ratio has stood, on average, at a negative 1,6 per cent of disposable income, slightly lower than in the previous five years. At the same time, from a supply-side point of view, South Africa is arguably distant from the ‘technological frontier’; even a slower rate of innovation in advanced economies should therefore not preclude faster domestic productivity growth, provided that a greater share of industrial sectors adopts more efficient and sophisticated technologies. Indeed, most upper-middle-income economies, with income-per-head levels broadly similar to South Africa’s, have experienced, on average, higher growth in GDP per capita in recent years.
Other explanations must therefore be sought for the domestic economy’s poor performance in recent years. A structural shortage of skills, which limits the economy’s ability to properly utilise its labour supply, has long been recognised and may have worsened over time amid ongoing poor educational outcomes.
In addition, infrastructure constraints, particularly in the electricity and transportation sectors, appear to have weighed on investment in new capacities, as they limit the viability and increase the logistics-related costs of new projects. Fortunately, progress is being made with addressing the particular bottleneck around electricity supply. Another important issue has been policy uncertainty about the tax and regulatory environment, which probably undermined expansion, in particular in sectors such as mining.
Empirical evidence that drivers of slower economic growth are different in South Africa from those in advanced economies may be found in the relative response of inflation. As I’ve mentioned earlier, inflation in the developed world has consistently fallen short of targets despite ample monetary stimulus, but it has remained elevated in South Africa.
Admittedly, domestic demand is weak, which has helped to limit the pass-through of rand depreciation, in recent years, to final inflation readings. Nonetheless, inflation expectations – as measured by the Bureau for Economic Research – have remained stuck at around the upper end of the SARB’s inflation target range, displaying no obvious sensitivity to slowing economic growth and contrasting with a declining trend observed in most advanced economies.
4. The rationale for a different monetary policy response
Let me now come to the issue of an appropriate monetary policy response to this new economic paradigm. Because South Africa has suffered a trend deceleration in growth (as did the advanced economies) and because central banks in the latter responded with aggressive stimulus, some people have been asking whether the SARB might not have been able to do more to support domestic activity.
However, as I have already highlighted, the drivers of weaker economic growth were different in South Africa, and many consequences differed too. One would therefore be mistaken in assuming that a ‘one size fits all’ policy response would also be appropriate for our domestic situation.
In the advanced economies, large-scale monetary easing and unconventional policy measures at first aimed to restore the proper functioning in money and asset markets, which was in jeopardy at the height of the Global Financial Crisis. At a later stage, these policies strove to facilitate the healing of stretched private-sector balance sheets while ensuring that both the supply of and the demand for credit continued to expand. In many ways, the key challenge was to prevent a broad-based ‘credit crunch’ which could have triggered widespread defaults, exacerbated the fragility of banks’ balance sheets, and threatened a debt-deflation spiral.
Such concerns were, however, far more muted in the case of South Africa. While they deteriorated somewhat around the time of the Global Financial Crisis, market perceptions of domestic banks’ credit risk never escalated to levels that would have threatened shortages of liquidity in the interbank market. In fact, banks remained adequately capitalised throughout the 2008/09 recession, and while the rates of non-performing loans increased significantly as economic activity contracted, this move was later reversed. Private-sector credit continued to grow, albeit at a lesser pace than prior to the recession, but with a notable reorientation of new lending from mortgages towards other forms of (mostly unsecured) credit.
This is not to say, however, that the SARB did not quickly respond to the rapidly changing economic conditions at the time. In fact, the central bank aggressively cut its repurchase rate from a pre-recession high of 12,00 per cent in December 2008 to 7,00 per cent nine months later. Compared with the previous interest rate cycles of the past 20 years, this was an unusually large move over an unusually short period of time. Further easing followed in 2010 and 2012, bringing the policy rate to 5,00 per cent, the lowest since the early 1970s.
Real forward-looking interest rates, obtained by subtracting broad-based inflation expectations two years ahead (as measured by the Bureau for Economic Research) from the nominal repurchase rate, remained in negative territory from mid-2009 to early 2015, indicative of a prolonged accommodative stance. Many economic indicators suggest that this loosening of monetary policy helped economic agents to deal with the most severe consequences of the recession – and hence limit both its length and magnitude.
For instance, South African households were able to reduce their leverage ratio without any major cutbacks in consumption. (Household debt expressed as a share of disposable income fell from a peak of 87,8 per cent in the first quarter of 2008 to 75,1 per cent in the second quarter of 2016.)
In fact, over the past five years, private consumption has grown by an annual average of 2,6 per cent, significantly exceeding the pace of real GDP growth. At the same time, the rate of home repossessions has edged consistently lower following a post-recession peak in 2011, and is now at very low levels. Insolvencies and liquidations have followed a similar path.
In other segments of the economy, though, monetary policy stimulus has had less of an impact, although evidence suggests that this was not due to an insufficiently low level of real interest rates. Of particular concern to the SARB has been the weakness in private fixed investment, which has declined, as a share of GDP, from 14,8 per cent in 2008 to 12,3 per cent in the second quarter of 2016.
If it continues, weakness in investment risks perpetuating the current phase of slow economic growth as it results in slower expansion and growing obsolescence of the capital stock, in turn also negatively impacting on potential growth. However, anecdotal evidence and surveys suggest that, rather than the level of borrowing costs, it is the level of uncertainties about future demand, the weaker price of commodities as well as political and regulatory uncertainty that are responsible for such weakness in fixed investment.
5. Why the South African Reserve Bank needed to raise interest rates
In early 2014, economic and market fundamentals came to a juncture that justified a tightening of the SARB’s monetary stance, lest inflation be allowed to rise to levels inconsistent with its price stability mandate.
The combination of weakening GDP growth, a widening current account deficit (under the joint effect of deteriorating terms of trade and the limited sensitivity of import and export volumes to real exchange rate depreciation) as well as low real domestic interest rates complicated South Africa’s external financing, resulting in a marked depreciation of the rand. In fact, the rand weakened not just against the world’s major currencies but also against ‘peer’ emerging market currencies and those of other commodity-exporting countries, suggesting a deterioration of international investor confidence in South Africa’s credit quality and macroeconomic fundamentals.
To avoid the rapid rand depreciation feeding into significantly higher inflation and leading to an unhooking of inflation expectations (which, as I’ve mentioned before, remained uncomfortably close to the top end of our target range), the SARB raised its repurchase rate by 200 basis points in six instalments between January 2014 and March 2016 to the current 7 per cent.
I must stress, however, that this pace of interest rate increases has been gradual compared with previous phases of policy tightening, for instance those initiated in 1998, 2001 and 2006. In addition, while the rate hikes have resulted in a reduced degree of monetary stimulus, it is difficult to argue that they have taken the monetary policy stance into outright restrictive territory: at around 1,0 per cent, the forward-looking real interest rate remains low compared with the average of the past 15-20 years.
The flexible nature of our inflation-targeting framework, which enables us to focus on medium-term projections for inflation rather than on the most recent developments in the CPI6, has allowed the central bank to raise rates in a gradual manner, taking into account the weakness of the domestic economy and the lack of demand-driven upward pressure on prices. It is worth repeating that, in the absence of demand-side pressures, the gradual hiking cycle was in response to our concern that repeated shocks would drive inflation persistently above the target range and pull with it inflation expectations and wage growth unmatched by productivity gains.
6. The way forward: the outlook for South Africa’s monetary policy
In light of the reduced degree of monetary stimulus, a debate has emerged in recent months on whether the SARB should implement additional rate hikes at all – and even whether we could afford to reverse part of the last two years’ hikes. This debate intensified towards the middle of August, building on a sustained recovery in the rand’s exchange rate from early 2016 lows and on several months of lower-than- 6 cosumer price index 10 expected inflation readings.
By the 23rd of August, the forward-rate agreement (FRA) curve had become almost entirely flat, suggesting that as many market participants saw room for a possible easing of policy as for additional rate hikes. Since that date, renewed fluctuations in the rand’s exchange rate, partly driven by investor perceptions of domestic risk, have altered the slope of the FRA curve. Nonetheless, how close we might be to the peak in the current interest rate cycle remains a key focus of investors. It would not surprise you that the MPC has similarly been seized with this matter, as policy decisions become rather finely balanced when one approaches possible turning points.
As indicated in previous policy statements, the MPC has felt in recent months that several developments, both global and domestic, have allowed it to press the pause button in the rate-hiking cycle. Since the start of 2016, monetary policies in large advanced economies have remained accommodative while the fears of a sizable devaluation of China’s currency have subsided and commodity prices have posted a moderate recovery.
Financial market volatility also subsided surprisingly quickly after the unexpected Brexit outcome (certainly relative to the severity of the underlying event and its possible consequences), and expectations of a policy adjustment in the US increasingly moved away from the Federal Reserve meeting in September.
This prompted renewed capital flows into emerging markets and allowed a recovery in the rand since the beginning of the year, even though short-term exchange rate volatility remains high. On the domestic front, inflation readings, especially for the core rate of CPI inflation, increased by a lesser amount than the SARB’s econometric models had projected earlier in the year. This may be a sign that the lower pass-through of foreign exchange depreciation to final inflation readings, which had been observed in the aftermath of the Global Financial Crisis, has persisted so far into the current cycle.
These developments have supported a lowering of the SARB’s projected trajectory for both headline and core CPI inflation. Headline inflation is now expected to return to within the target range earlier than previously expected, and with a lower peak in the fourth quarter of this year, while the core measure is no longer seen breaching 6,0 per cent.
Despite the improvement in the balance on the current account in the second quarter of the year, South Africa is likely to continue running significant, structural current account deficits in the near- to medium-term future, owing in part to our reliance on imported capital goods for domestic infrastructure needs and the challenges of our mining industry.
In light of this reliance on foreign capital, the South African rand remains at risk of sudden changes in global investor sentiment. Other factors pose an upside risk to inflation, including the possibility of faster wage settlements as unions seek offsets for a quicker pace of headline CPI increases and the odds that a persisting drought could keep food prices higher than currently projected.
Finally, there is no certainty that the pass-through from rand depreciation – whose dynamics we are yet to fully understand – will remain as low as it has been in recent years. At the same time, though, inflation could surprise on the downside, for instance if the recent upward momentum in the rand’s exchange rate is sustained or if increased rainfall triggers a marked drop in grain prices.
At this stage, the MPC views the risks to the most recent forecast to be more or less balanced; it has therefore indicated that, in light of the new forecast, we may be close to the end of the tightening cycle. However, such ‘guidance’ remains conditional on the absence of the risks highlighted materialising in the near future.
Indeed, there is no guarantee that the causes of the reduced risks to the inflation outlook will persist in coming quarters, and the MPC has been careful to convey that some of the factors which have had a favourable impact on the inflation outlook could reverse quickly, in which case the view that we are close to the end of the tightening cycle would need to be reassessed. The MPC has also indicated that the bar to interest rate cuts is high. More than ever, policy must remain dependent on forthcoming data and events, and continued vigilance is warranted.
7. Conclusion
In conclusion, let me reiterate that the SARB remains committed to achieving its mandate of price stability, but it acts in a flexible manner, taking into account the real cycle of the domestic economy and the way in which it affects the price formation process. It is our belief that this flexible approach has allowed the economy to better withstand the recent shocks from global developments and the end of the commodity price ‘super-cycle’ while allowing us to fulfil our mandate of contributing to financial stability.
However, I must also stress that there are limitations to how far the SARB – or any other central bank, for that matter – can support economic activity. Because of the unique nature of their mandates, there are only a limited number of tools that central banks, including ours, have at their disposal.
Over the last few years – in a global environment where elevated public debt levels have constrained the use and potential effectiveness of fiscal policy and where voter discontent often makes policymakers shy of implementing structural reforms – the tendency has grown to view central banks as ‘the policymakers of last resort’, with a prime responsibility for rekindling economic growth.
In the words of well-known commentator Mohamed El-Erian, they are increasingly seen as ‘the only game in town’. While the allocating of a disproportionate burden to central banks should be avoided, their responsibilities are certainly not to be minimised: low and stable inflation, together with ongoing financial stability and a proper functioning of financial markets, is a necessary precondition for sustained, long-term, inclusive economic growth.
Thank you!
