/ 10 September 2010

Productivity is our best currency

A weak rand is unlikely to help South African manufacturing, argues Adrian Saville

The South African manufacturing sector has been in a state of decline for more than a decade (Dube, Hausmann and Rodrik, 2007). While it is arguable that the decline is structural in nature, recently it has become popular to appeal to policymakers to intervene in foreign exchange markets to bring about a weakening in the rand against the currencies of South Africa’s major trading partners. Such rand-weakness advocates argue that a depreciation in the rand will provide relief for a domestic manufacturing sector under acute pressure.

Since 2008 calls for policy intervention to bring about a weakening in the rand have become shrill as a consequence of a number of negative factors that have emerged out of the global financial crisis. In particular, two factors stand out, namely weak global economic performance in the past two years, aggravated by strength in the rand against the so-called “hard” currencies of our major trading partners.

Regardless of the detail, two themes emerge from the evidence, namely: (a) the particular stress that the manufacturing sector has come under in the past two years, which has translated into falling production and employment; and (b) the rand strength that has come about over the same period.

In turn, a number of commentators have identified strength in the rand as a key factor that explains the decline in the manufacturing sector. And, in turn, a growing voice has emerged out of the domestic landscape calling for policy interventions that will precipitate a weakening in the rand and, in so doing, help restore the competitiveness of South African manufactured exports in global markets.

As an aside, it is interesting to note that the appeal for rand weakness enjoys widespread support from a diverse set of interests, which includes the Organisation for Economic Cooperation and Development; the Harvard Panel; domestic labour federations, including the Congress of South African Trade Unions (Cosatu); and firms that are part of South Africa’s manufacturing sector. However, widespread support is not always indicative of sensible policy. In fact, based on evidence drawn from a study of domestic manufacturing sector data, it seems that there is little basis for the argument that a weak (or weaker) rand will help South African manufacturers compete in international markets or, for that matter, domestically.

An examination of the data that capture the relationship between the performance of the manufacturing sector and movements in the rand in the past three decades reveals a result that runs contrary to conventional wisdom and the arguments of mainstream economic textbooks. Specifically, the evidence in the South African economy shows no explanatory relationship between a depreciating rand and growth in the manufacturing sector.

The results of the findings mean that, in the past 30 years, a strengthening of the rand corresponds with growth in the manufacturing sector, whereas the sector declines if the rand weakens. This is the exact opposite of what textbook economics and conventional argument would have us expect.

Even by further mining the data, for instance by selecting different data sets — like shorter time periods that exclude the politically volatile 1980s, or times of extreme rand weakness, such as end-2001 into early-2002 — no significant correlation between the two variables is observed.

A possible explanation for this result, which based on our observations seems to be broadly unanticipated, is that South African manufacturers participate in global economic growth when the world’s economies are performing well. However, this rise in sales is not caused by a gain in market share (a proxy for competitiveness). Rather, sales are lifted by the rising tide of world economic growth. The graph showing exports from South Africa and G7 GDP provides powerful anecdotal support for this argument. At the same time the more buoyant global conditions convert into direct and portfolio investment flows into emerging markets, which includes an appetite for South African assets. In turn, the sentiment favouring South African assets translates into a strengthening rand.

In addition to the key result that rand weakness and manufacturing growth appear to be inversely related, it is worth noting that other problems exist with the weak-rand argument.

First, even if we were to find strong evidence to support the weak-rand hypothesis, the argument then requires that policymakers are in a position to control (or at least heavily influence) the level of the rand. The average daily turnover on the South African foreign exchange market is in the order of $10-billion (R72,6-billion). This compares with the size of foreign exchange reserves held by the South African Reserve Bank of just more than $40-billion, equivalent to four days’ turnover on the foreign exchange market.

Second, if the weak-rand hypothesis held up, who gets to choose the “right” level for the rand? Further, what happens when that right level is no longer right? The risk of once-off intervention converting into regular price-fixing becomes apparent.

Third, it is arguably perverse that South Africa has foreign exchange controls designed to keep capital in the country while, at the same time, there is a growing body that calls for the rand to be allowed to go weaker by, for example, taxing inflows through a Tobin tax.

Fourth, “weak randers” appear to have confused South Africa’s income statement and balance sheet, or simply look straight past the balance sheet effect of a weaker rand.

If the rand were allowed to weaken, say to R14 to the dollar, this might give our manufacturers a boost to get into markets and in so doing help with market share. But the cost of this, among other things, would be to devastate the country’s balance sheet. In the current example, the country’s net assets (equity) measured in global purchasing power terms, by definition, would be halved. Boosting the income statement of a firm or a country by deliberately weakening the balance sheet is not a smart way to compete, neither is it sustainable. In fact, if it worked, it would quickly translate into a “race to the bottom” among firms and countries hungry for export market share.

The above arguments and evidence point to the fact that debating an ideal value for the currency is debating the wrong thing. The success of South African manufacturing and its global competitiveness resides elsewhere.

To this end, it equips us to recognise that labour is the single most important input into manufacturing. By extension, labour productivity is the single most important factor in determining South Africa’s relative competitiveness.

While productivity of capital and total factor productivity cannot be overlooked, it is labour productivity that provides the key to global competitiveness. Importantly, the labour productivity argument is not about mimicking the Chinese model of promoting competitiveness by way of low wages. Indeed, low wages mean low per capita incomes which, by definition, cannot translate into economic success for individuals or the country.

However, by focusing on promoting productivity and sharing gains in productivity, South African manufacturers will establish a source of sustained competitive gains, while labour will establish a sustainable source of income growth.

In time the weak-rand advocates will be surprised to find that this world of higher productivity and higher incomes correlates with a stronger rand, greater global competitiveness of South African firms and higher standards of living for South Africans generally.

Adrian Saville is the chief investment officer at Cannon Asset Managers.This is an edited version of a paper he wrote titled “Cents and Sensibility”. To contribute to the debate on trade policy and the rand email [email protected]