SA economy: The rand, exports and interest rates

I wanted to start my throwing around of ideas about the South African economy with a simpler story of the drivers for growth, but this week the depreciation of the exchange rate and repo rate increase by the Reserve Bank makes for the big news.

One of the stories is that the weak rand will help to boost exports. Mike Schussler ripped into this idea earlier but also talked about a lot of the other challenges facing the SA economy. Depreciation of the rand cannot be seen as a growth strategy.

To add some academic perspective I wrote a post on the School blog drawing on a paper by Lawrence Edwards and Robert Garlic. The heart of the story is the following:

At an aggregate level there is a positive association between exchange rate depreciations and export performance. Edwards and Garlic lists the results from a range of studies and show that “a one percent rand depreciation is estimated to raise long-run manufacturing exports by 0.78 to 1.38 percent”. But this aggregate relationship fails to capture changes in competitiveness at the sectoral level. Their evidence shows that primary products are less responsive to exchange rate shocks than manufactured products. Non-gold merchandise exports (including manufacturing) are less responsive to exchange rate shocks than manufacturing alone. There are also differences across manufacturing sectors: the export response in natural resource-based and machinery & metal products sectors is generally lower than in labour intensive, chemical-intensive beneficiated sectors. And what about the channels through which the exchange rate affects exports? The evidence shows that South African exporters of manufactured goods are price takers in the international market. That means that export growth is not constrained by inelastic foreign demand or an inability to price competitively in the international market. This sound like promising news the minister would like to hear, but as always there are caveats: 

• The extent of an export boom depends on the composition of exports. Historical evidence showed that manufactured exports are likely to benefit more than commodity exports, but looking at the nature of South Africa’s top-20 exports (from Edwards & Alves in SAJE, 2006) one sees that manufactured exports make up a small part of the mix.
• The length of an export boom depends on competitiveness, specifically on the cost of imported inputs and labour. The evidence shows that nominal depreciations have not sustained the profitability of exports. Domestic producer prices are very responsive to changes in the exchange rate – a 1 percent rise in import prices has raised domestic producer prices by between 0.85 and 1 percent in the long run.

Even if exporters are winners for a while, the South African consumer is a clear loser when the rand depreciates. Yesterday Neil Rankin tweeted a link to a paper that he wrote with Aron, Creamer and Muellbauer on the exchange rate pass-through to consumer prices. They found that:

overall pass-through to the almost 63 per cent of the CPI covered is about 30 per cent after two years, and higher for food.

Finally, I found this nice tweet on research about the use of interest rates to defend a currency. It is in the context of fixed exchange rates and in the South African case we are not really defending the rand, but rather the current account and fighting inflation expectations, but it still offers an interesting take-away: higher interest rates helps a bit, but only for a while.

Do High Interest Rates Defend Currencies During Speculative Attacks? @tylercowen http://t.co/tR1XudFNny http://t.co/xXuPANg2b6
— MacroDigest (@Macro_Digest) January 29, 2014

Behavioral economics

I found this cool post at Marginal Economics over the weekend. It describes the Busara Centre of Behavioral Economics and what they offer development economics researchers through experiments in their lab. Now I'm just mailing this idea to myself: we need a lab like this for firm-level research in South Africa. Surveys are expensive and time consuming and we may also learn a lot about firms and entrepreneurs in an experimental setup.I know that Prof Wim Naudé at MsM has been thinking about this too.

Does anyone out there know more about firm-level research and the ways that experiments have been used to examine questions related to entrepreneurship, risk-return, deciding to export or to hire more people?

ERSA workshop

This week I am attending the ERSA workshop Contemporary Analysis of International Trade. The workshop is built around what Jim Fairburn calls FUGSA - the Feenstra User Group SA - and Jim, Lawrence Edwards and Neil Rankin have been covering a number of topics in International Trade based on Feenstra's book.

 

On Monday Jim discussed the link between trade and growth in three parts:

• The links between trade and productivity,
• Growth and monopolistic competition, and
• The empirical evidence on trade and growth.

Neil Rankin followed with a presentation on models and evidence on firms and international trade. The key framework is the Mellitz model and building on an explanation of the model, Neil presented some interesting points:

• Few firms export and exporters are diffferent - they are typically large, established firms that are capital intensive and use high-skill labour, they are more productive and pay higher wages.
• Multi-product firms that exporting to multiple countries are rare, but they contribute the bulk of the value of exports.
• There are more single-product firms exporting to a single country but the values are low.
• Exporters are significant importers of inputs.

This simplifies a much longer and involved explanation of firms and trade, but clearly raises some interesting issues:

• Is it self-selection or can exporters learn by doing? Who are the so-called "born global" firms?
• Looking at U.S. evidence it does not seem to help to aid or develop small-firm exporters. Policy should rather support firms to grow big and some will export.
• If exporters import inputs it does not seem to be sensible to weaken the exchange rate to support exporters.

Today saw presentations on anti-dumping, micro-evidence on price setting and product market evidence of African integration, but I accidentally deleted my UPAD notes and won't comment off the cuff anymore than saying it was interesting. If the slides are shared, I'll add the stories!

Megapolitan regions

This week has seen some interesting news coverage of a new book Megapolitan America by Arthur Nelson and Robert Lang. The Atlantic Cities showed a map of large regions of interconnected metropolitan areas.

These megapolitan regions are expected to house two thirds of the U.S population by 2040 and are seen as the interface with the global economy. Such agglomerations drive growth through the external economies that they generate. Economic geographers argue that concentration of economic activity cumulatively causes a thick labour market, specialised suppliers of intermediate inputs and knowledge spillovers 'that are in the air'. The end result is lower costs and faster growth.

The coverage by The Atlantic Cities focussed on the role that infrastructure can play in supporting such agglomerations. The Dallas/Fort Worth International Airport is presented as a success in coordination of infrastructure investment between metro regions.

In South Africa drawing a map like this of economic activity (local GDP) is complicated and the topic of a lot of my data work planned for this year. It is slightly easier to track down exporters through SARS data and work by two of my colleagues (Wim Naudé and Marianne Matthee) show interesting agglomerations.

Exporters typically require the thick labour markets, suppliers of intermediates, knowledge spillovers and infrastructure characteristic of agglomerations. In South Africa the port cities of Cape Town, Port Elizabeth and Durban are key agglomerations but the landlocked Gauteng economy is the powerhouse.

Hopefully I will be taking a closer look at South Africa agglomerations in the year to come. Stay tuned...