Lesetja Kganyago: Inflation, but no jobs or growth - policy responses to South Africa's economic challenges

Address by Mr Lesetja Kganyago, Governor of the South African Reserve Bank, at the Labour Law Conference, Johannesburg, 24 August 2016.

The views expressed in this speech are those of the speaker and not the view of the BIS.

Central bank speech  | 
25 August 2016
PDF version
 |  7 pages

Good morning, ladies and gentlemen

Thank you for inviting me here today to address this forum.

South Africa is suffering from a combination of stagnant economic growth and rising inflation. The economy contracted in the first quarter of the year, and although the second quarter looks considerably better, we expect no growth in 2016 overall - which compares unfavourably to the 1,3 per cent achieved in 2015 or the 3 per cent average since 1994. Furthermore, for the first time since the global financial crisis, we expect to lose jobs this year, on a net basis. Yet even as economic growth and job creation stall, prices - fuelled by food price and currency shocks - as well as wages and salaries continue to increase above the targeted inflation rate.1 This keeps headline inflation higher than economic conditions warrant. Underlying this inflation outcome is a set of markets, rules and systems that generate strong price, wage and salary pressures and play a large role in generating this combination of high inflation, high unemployment, and low growth.

Today, I will discuss policy options for addressing this problem, including the contribution from monetary policy.

One of the things we need is for price and wage demands to abide by the inflation target, so that we achieve a lower inflation rate more consistently. Attaining these outcomes will require change, including institutional reforms. And reforms will create space for monetary policy. We would prefer not to tighten policy during a slowdown, but as long as our economy is geared to produce weak growth and high inflation, we have to do just that. Higher inflation begets higher interest rates. Lower inflation, alternatively, enhances the competitiveness of the economy and reduces interest rates, supporting investment, job creation, and poverty reduction.

South Africa as a labour market outlier

South Africa's unemployment rates, narrowly defined, have exceeded 20 per cent since the early 1990s. From a comparative perspective, sustained unemployment rates at these levels are very rare. At the moment, according to data published by the International Monetary Fund (IMF), they are shared by just four other countries. Two of these are Greece and Spain, which have suffered intense and prolonged economic crises.2 But both expect to see unemployment decline in the next few years. In fact, Spanish data in May showed unemployment falling to just under 20 per cent for the first time since the global financial crisis. In South Africa, no such improvement is being forecast. In fact, the IMF numbers show our unemployment deteriorating, from about 25 per cent now to around 27 per cent in three years' time.

It is quite extraordinary for unemployment to remain so high throughout an economic cycle. Compare South Africa with Brazil, a country similar in many ways. When Brazil's economy boomed, unemployment fell to below 5 per cent. Since then, Brazil has fallen into a deep recession, with growth close to minus 4 per cent in each of the past two years, the worst performance in a century, including the Great Depression.

Unemployment has more than doubled; it is now over 10 per cent. This is a terrible economic experience. But if we could get unemployment in South Africa to 10 per cent, that would be viewed as a tremendous success. There isn't much to envy about Brazil's economy these days, but I would happily take their unemployment rates.

Many observers look at South Africa's levels of joblessness and reflexively describe them as unsustainable. But they have persisted for around two decades, so it is a bold prediction to say that they cannot continue. A better interpretation is that we have very unusual, even unnatural, labour market dynamics. This implies that we have vast scope for improvement. Almost every other country has a larger share of its people in work - and most have a larger share participating in the workforce.

The National Development Plan sets unemployment goals of 14 per cent by 2020 and 6 per cent by 2030. For labour force participation rates, the plan envisions 60 per cent by 2020 and 65 per cent by 2030, up from the estimated 56 per cent today.3 We don't have to achieve a world-beating performance to arrive at these sorts of numbers. Mexico has a 65 per cent labour force participation rate right now. Brazil is at 70 per cent. Meanwhile, world average unemployment rates are at about 5,5 per cent currently, and a little lower in emerging markets.4 In other words, the goals of the National Development Plan are perfectly sensible. By world standards, they are normal, or even a little worse than normal. But if we're going to get there, we're going to need to create 11 million new jobs given the population growth.

At the moment, however, we are going in the wrong direction. Both the Quarterly Employment Statistics and the Quarterly Labour Force Survey have recently shown declines in the number of people working - the first negative numbers since the global financial crisis. Our forecasts suggest that this trend will continue, as the public sector achieves financial sustainability and the private sector only slowly recovers from the current lows.

The private sector remains the major employer in this economy, providing around three-quarters of all jobs. And while we have seen higher employment in service sectors such as finance and trade, catering, and accommodation since the crisis, private-sector job growth stalled a few years ago. The mining sector is grappling with lower commodity prices and a difficult investment environment; it now employs about 80 000 fewer people than it did at its peak in mid-2012. The manufacturing sector continues to shed workers, persisting on a multi-decade trend. We need to get private-sector employment going again.

The public sector employs far fewer people than the private sector does, but it has been the prime source of new jobs in the post-crisis period. Since 2009, public-sector employment has grown by about 300 000 people. The share of the budget spent on compensation has also risen markedly; South Africa's government wage bill is now one of the highest among emerging markets - not so much because the state employs an unusual number of people, but because it pays them comparatively well.5 Our national finances can no longer bear this sort of wage growth and government has had to implement a personnel freeze, removing one source of job growth in recent years.

The overall result is a bleak outlook for labour markets. The question is what to do about it.

The inclusive reform agenda

The answer is fundamentally about growing again, reforming product markets, and making labour markets more inclusive. We've built an economy that produces much more inflation than growth and spits out unemployment as a by-product. As long as we can use commodity prices or debt to grow - the path taken between 2004 and 2014 - we look like we're making progress. The economy expands (albeit modestly) and jobs are sustained, but real income per capita doesn't move much.6 When the debt burdens get too heavy and commodity prices fall, however, inflation stays sticky while growth falls below the population growth rate, entrenching inequality and worsening poverty. At its core, the formal sector of the economy protects itself from the growth slowdown by raising prices, aggravating the downturn, and imposing the cost on those losing their jobs and those without the market power to set prices and wages.

There is an alternative path, in which strong investment and increasing labour force absorption allow the economy to expand without accelerating inflation.7 That's where we need to go. There are a series of product and labour market reforms that can help us get there.

Start with product markets, which offer us our most attractive reform opportunities. OECD8 studies of product market regulation show that South Africa has unusually high barriers to entry for new businesses.9 Targeting and reducing these barriers would encourage new business creation and investment, and would raise productivity. In particular, lowering the costs in network industries - such as telecommunications, energy, and transport - would benefit many other parts of the economy. The cost of utilities feeds into the underlying cost structure of the economy; it makes South African goods expensive for the export market and raises that cost of living for the average South African. Research suggests that we could raise our potential growth rate by 1,5 percentage points by reducing the markups in product markets to world benchmarks - which would roughly double the economy's current "speed limit".10

Of course, structural reforms of this type are difficult. Their benefits, in the form of stronger growth and more employment, tend to be widely distributed across the population. The diffusion of benefits is important in the policy debate, because the costs, by contrast, generally appear quickly and affect a smaller number of people. For example, bringing down the price of a product like electricity is good for all the millions of users of electricity but painful for the producer, the latter having to become more efficient. For this reason, reform opponents usually mobilise more effectively than the prospective beneficiaries.

Nonetheless, there are extra, easy benefits that make reform yet more attractive right now. South Africa has acquired a reputation for devising sound development strategies and then failing to implement them. Following through on reform could reframe the whole South African narrative and boost business confidence. Global conditions would also help. Interest rates are now exceptionally low internationally, and investors are desperately searching for yield. That's an opportunity; a lot of money would move this way if we became a better growth and investment story.

Then there are reforms which affect the labour market directly. South Africa, arguably, has a three-tier labour market. One is mostly informal and pays low wages. The second is formalised and pays intermediate-range salaries, typically negotiated on an industry-wide basis. The third tier pays the highest salaries and features mostly highly skilled employees. Each of these poses different challenges.

On average, participants in the third market earn the highest salaries and benefit from skills shortages in South Africa. These shortages mean that employer competition for workers is quite intense, and unemployment in this category is therefore relatively low. But this doesn't mean that the labour market for the highly skilled is a healthy one. The elevated wages in this market feed our high levels of inequality and contribute to inflation. Furthermore, skills shortages are a problem for economic growth and employment creation. Workers of different skill levels are complementary "goods"; with more engineers you can employ more construction workers to build more bridges. What this segment of the labour market needs, in the longer run, is more and better education to open the highly skilled job market to more South Africans and better match what students study with the skills required by the private sector. We could also temporarily welcome more skilled immigrants to increase competition and help firms find the people they need to grow while we invest in growing our domestic knowledge base. The possibility of pro-growth, pro-equality reform in this area should make it a policy priority.

South Africa's other two labour markets don't absorb nearly enough people, which leaves many South Africans unemployed or out of the labour force entirely. As the National Development Plan recognises, this problem has a lot to do with spatial legacies and access to information about work. As a result, the plan pays special attention to the distances between people and jobs. It also points out that if we are going to create 11 million new jobs by 2030, many of them cannot be highly paid. This ties in with the product market reform agenda: if the wages in new jobs can't be that high, we need to achieve downward pressure on prices so that the possible wages command decent buying power. It also speaks to the links between wages and employment. As Bheki Ntshalintshali of COSATU11 recently reminded us, there is the real risk of excessive wage demands producing higher unemployment.12

We know that the agricultural sector shed hundreds of thousands of workers after the imposition of a minimum wage at an excessive level.13 We know that this economy lost a disproportionate number of jobs during the global financial crisis relative to the decline in economic output, in large part because of high wages and rigidities during periods of normal business operations.14 Based on historical trends, we forecast persistent wage growth despite rising unemployment and minimal growth. If we try to play a game in which fewer and fewer people get higher and higher wages while more and more people wind up unemployed, we can only lose. We have to get more people into work.

To tackle this problem, one of the things we should do is reconsider the institutions we use to negotiate wage pricing. There is an important argument in academic literature, specifically by Lars Calmfors and John Driffill, connecting unemployment to the kinds of collective bargaining rules used in a country.15 The basic idea is that a country can achieve better employment outcomes when collective bargaining is done either at a low level, such as one union per firm or plant, or at a very high level, such as the whole economy. The worst outcome is to be in the middle, between the two options. This is because company or plant-specific unions have a special interest in the health of the firm that employs their members and will not want agreements that make the firm unsustainable. By contrast, very large unions have the power to shape macroeconomic outcomes. They will therefore avoid making deals which will push up inflation, drive up interest rates, reduce investment and growth, and ultimately destroy jobs.

By contrast, medium-sized institutions, such as those that cover industries, are big enough to inflict costs on the economy but not big enough to experience those costs themselves - the costs are pushed onto others. In these systems, it is easy to reach a deal that keeps out new and smaller firms, so you get a lower level of employment at higher wages and a smaller number of businesses with larger profits. This is a reasonable description of South Africa's collective bargaining arrangements. We should not be surprised that this is contributing to bad employment outcomes, an effect probably only partially mitigated by exemptions.16 Indeed, there are reliable estimates suggesting that these arrangements reduce employment in affected industries by 8-13 per cent, in large part because they hurt small, labour-intensive firms.17 South Africa has surprisingly few of these firms, relative to its peers. This helps explain our unusually high levels of unemployment.

The reform conversation and the problem of good faith

Ladies and gentlemen, labour market discussions in South Africa are difficult. In particular, they are difficult because they are convened against a backdrop of historical oppression and deep race and class divides.

As in some other countries with large economic and social problems, the discussion quickly moves beyond economics. It stops being about appropriate trade-offs and the sorts of experiments which might get us to a better place. Rather, it becomes a discussion about fear of the future and about a shortage of trust; the exchange stops too soon and often at the point where insults fly. If the conversation ends there, nothing gets solved. We are, together, responsible for the choices we make now - and we have the power to make better choices.

The people who have the most to gain from reform are the people locked out of the economy. They don't have much power. Government therefore has a special duty to prioritise their interests. For many South Africans in jobs, the struggle is to hold on to employment and keep wages growing enough to support large numbers of dependents. We need to turn this around, increasing employment and lowering dependency ratios. The best way to raise household incomes is to increase employment. A growing economy and rising productivity will push up real incomes for workers; this need not be done at the expense of jobs. For this reason, there will be long-run benefits for both the unemployed and the workers in the reform agenda.

By contrast, opening up product markets will deprive some people of lucrative, over-protected niches in the economy. That's a good thing. Capitalists often preach the benefits of free markets; we think they should try some competition themselves. We can also see that the labour market reforms needed to temper bad economic outcomes should go hand in hand with other reforms to boost economic growth. By achieving both sets of reforms, we will create permanent gains from greater inclusion and higher productivity - gains sufficient to offset any temporary economic sacrifices imposed on our economy's "insiders".

In addition to managing labour laws, government also has a role to play in managing labour market dynamics. Firstly, by writing the rules and regulations which ensure that product markets remain competitive. Secondly, by providing basic services that are accessible and lower the cost of living. Planning for affordable housing that is closer to centres of employment and the provision of adequate public transport systems would be examples of these. One of the things that have become clear from the experience in the mining sector is that we do not have adequate rental housing stock for entry-level workers. Thirdly, government will support long-term labour market productivity by providing quality education and health care. Finally, government can ensure that policy conversations balance the interests of "insiders" and "outsiders".

The role of monetary policy

I'd like to end on the subject of the contribution from monetary policy.

The South African Reserve Bank has been tasked with maintaining inflation within the target range of 3-6 per cent, to be implemented flexibly. Being flexible means that inflation may, from time to time, find itself temporarily outside the target range. There are several reasons for this. Sometimes, there are inflation surprises and it is impossible to change policy in time to have any impact. Sometimes, it would be costly to return inflation to target very rapidly, so we may prefer to disinflate more gradually. And sometimes, we choose to look through departures from the target range because they are expected to be temporary.

But there are limits to how flexible we can be. We acknowledge that inflation has been rising recently in response to supply shocks. South Africa has been suffering from a severe drought, which has so far pushed food prices 11 per cent higher than they were in the middle of last year. We are also seeing the effects of sustained currency depreciation, with the rand losing about half its value against the US dollar over the period May 2013 to July 2016.

Our forecasts, and our experience, suggest that shocks of this scale pass through into prices and wages more generally and over time, accelerating both inflation and unemployment. This would have been an unacceptable outcome. Accordingly, we have been raising interest rates in a gradual way, taking the repo rate from 5,0 per cent in January 2014 to 7,0 per cent in March this year. By defending the target range, even in difficult conditions, we have helped to control the expectations of future inflation. Together with some currency strength, and the prospect of softening food prices, we expect underlying inflation to remain reasonably well contained. And yet, with inflation hovering at around 6,5 per cent, the real interest rate level of about 0,5 per cent remains exceptionally supportive of economic activity.

We are sometimes asked how we could have raised interest rates when the economy was slowing, but when inflation is sticky and even rising despite slowing growth, more and more monetary accommodation has adverse longer-run consequences. If above-target inflation becomes the "new normal", longer-term interest rates will quickly rise in response. Lenders expect to be compensated for inflation and inflation risk. In turn, the share of current income that we need to spend on servicing debt increases, draining resources from other priorities, like investment.

So it is clear that we need to focus on doing things that reduce the general level of lending rates instead of permanently raising them. As I have discussed today, the markets for goods and services in South Africa are commonly dominated by a few big firms. And labour markets are uncompetitive because of skills shortages, restricted entry, and extended industry or sectoral determinations. The result is that prices and wages are only flexible in an upward direction, and are otherwise rigid.

This rigidity problem constrains our efforts to improve outcomes and permanently lower interest rates. Our inflation outcomes appear to have settled at around the top of the target range, close to 6 per cent. This implies that nominal market interest rates will never fall much below 9 per cent. This cannot be an acceptable end point of macroeconomic policy, but it also cannot be changed easily with shifts in the monetary policy stance. If we want lower interest rates and we want them without going through a recession, then we have to lower inflation outcomes arising from administered price processes and, above all else, the way in which prices and wages are determined in our collective bargaining system. Simply reducing the inflationary impulse from price and wage setting, to something around the middle of the inflation target range, would help to create jobs economy-wide and ease constraints on policy.


In conclusion, I have spoken today about some key challenges facing the South African economy. The solution probably isn't going to emerge from rising commodity prices or a big rebound in world growth which helps exports. It also cannot come from yet more fiscal or monetary stimulus. Macroeconomic policy is already doing what it can, but the constraints are tightening; we are running out of room to borrow and we have too much inflation to lower interest rates.

This leaves us with the option of reforms. South African product markets are ripe for change; opening up network industries would in particular have widespread growth benefits. Labour markets for skilled people could be made more competitive, helping growth and reducing inequality at the same time. Our wage bargaining institutions could be more supportive of better economic outcomes. Less nominal wage pressure on prices would help us to lower inflation and the overall level of interest rates. Our institutions could also be less hostile to small firms, the kinds of companies which create lots of jobs - and which are also scarcer here than they are in other countries. Getting more people into jobs would lower dependency ratios, reducing the urgency of high real wages which themselves price many people out of work entirely.

Finally, we need to think about the kind of conversation we are having. The reform discussion can be difficult, partly because of vested interests and partly because of a shortage of trust between parties. Given our difficult situation, we need to find ways of talking to each other that ultimately solve problems instead of merely emphasising our differences. We are all partners in this country and its future. We have a shared responsibility to make that future better.

Thank you.

1  Over the past five years, average wages and salaries have consistently outpaced inflation. From 2011 to 2015, annual headline inflation stood at 5 per cent, 5,7 per cent, 5,8 per cent, 6,1 per cent and 4,6 per cent. Average wages and salaries, by contrast, grew by 7,6 per cent, 7,5 per cent, 9,3 per cent, 7,7 per cent and 8,5 per cent over the same period. Employment growth has slowed every year from 2011 to 2015.

2  The other two are Bosnia and Herzegovina as well as the former Yugoslav Republic of Macedonia. These are more complicated cases than Greece or Spain, but both were affected by the violent breakup of Yugoslavia and the loss of historical markets.

3  National Planning Commission, Our future - make it work: National Development Plan - 2030, p. 118.

The labour force participation data are drawn from the Organisation for Economic Co-operation and Development database; see https://data.oecd.org/emp/labour-force-participation-rate.htm. The unemployment numbers are from the Bank for International Settlements; see http://www.bis.org/statistics/ar2016stats.htm, Figure I.1.

5 The average wage in the public sector is about 1,8 times the South African gross domestic product (GDP) per capita, well above the emerging market average (which is about 1,4 times the GDP per capita). See the International Monetary Fund, 2016 Article IV consultation: South Africa, pp. 18-20, July 2016 (https://www.imf.org/external/pubs/ft/scr/2016/cr16217.pdf).

6  From 1991 to 2000, the gross domestic product (GDP) per capita contracted slightly by, on average, -0,2 per cent per year. From 2001 to 2010, it grew by 2,2 per cent per year (on average). From 2011 to 2015, GDP growth per capita slowed to 0,5 per cent on average; the numbers for 2014 and 2015 are -0,04 per cent and -0,48 per cent respectively.

7  This is discussed, for instance, by Ruchir Sharma in The rise and fall of nations: forces of change in the post-crisis world, published by W.W. Norton & Company, New York, in 2016. See especially chapter 7.

Organisation for Economic Co-operation and Development.

9  This is available from the Organisation for Economic Co-operation and Development, Indicators of product market regulation (http://www.oecd.org/eco/growth/indicatorsofproductmarketregulationhomepage.htm).

10  See David Faulkner, Chris Loewald and Konstantin Makrelov, 'Achieving higher growth and employment: policy options for South Africa', Working Paper 13/03, July 2013 (http://www.resbank.co.za/Lists/News%20and%20Publications/Attachments/5806/WP1303.pdf).

11  Congress of South African Trade Unions.

12 'COSATU urges members to balance pay with job security in wage talks, Business Day, 4 July 2016 (http://www.bdlive.co.za/national/labour/2016/07/04/cosatu-urges-members-to-balance-pay-with-job-security-in-wage-talks).

13  See Haroon Bhorat, Ravi Kanbur and Benjamin Stanwix, 'Estimating the impact of minimum wages on employment, wages, and non-wage benefits: the case of agriculture in South Africa', American Journal of Agricultural Economics, first published online on 27 June 2014.

14 See Nir Klein, 'Real wage, labor productivity, and employment trends in South Africa: a closer look', International Monetary Fund Working Paper 12/92, April 2012.

15  See Lars Calmfors and John Driffill, 'Bargaining structure, corporatism and macroeconomic performance', Economic Policy, No. 88, 1988.

16  Bargaining council agreements typically disallow exemptions for "economic reasons". This presumably includes low productivity or unaffordability.

17  See Jeremy R. Magruder, 'High unemployment yet few small firms: the role of centralised bargaining in South Africa', American Economic Journal: Applied Economics, Volume 4, No. 3, July 2012.