Latin America
Dollar Pain Brings Latin Gain?
April 24, 2007

By Gray Newman | New York

After more than five years of decline, the recent drop in the dollar has set off concerns that a more pronounced fall may be in the offing.  Our currency economist Stephen Jen expects further dollar weakness, although he contends that the move downward is likely to be cyclical rather than structural (see “No Need to Look Under the Rocks for Reasons”, Global Economic Forum, April 20, 2007).  Meanwhile, many argue that a weaker dollar is necessary to help reduce the large US current account deficit.  Whether this year’s decline is cyclical or structural, the dollar is approaching levels not seen in nearly a decade, and that has investors asking what the implications of a further dollar move could be for emerging markets.  Here are our first thoughts on the implications of a weaker dollar on Latin America.

It’s not the fall that hurts…

First, and most obvious, an abrupt move downward in the dollar — a dollar crash — would almost certainly be negative for the region.  Risk appetite would likely shrink and Latin America would suffer in the first instance.  Whatever medium-term macro benefits that may exist thanks to a weaker dollar, we doubt that they would do much to save the region from a financial crisis in the US and indeed in global markets.

We have argued in recent years about the transformation in Latin America.  Debt levels in the region are down and international reserves are up with the emergence of current account surpluses.  Brazil, the poster child of the dual risks of capital controls and debt default just a little over four years ago, now has reserves piling up far in excess of its gross public external debt. But none of these improvements in the fundamentals seemed to matter much in the first days of last May’s panic.  Had the market nervousness lasted a bit longer last year, we could have been faced with major dislocations from some of the who’s who of emerging markets.  Indeed, I was struck by the naïveté of some emerging market policy makers last year, who thought that simply hiking interest rates would return order.  The damage to growth and debt sustainability could have quickly overwhelmed the textbook monetary response had market nervousness not abated. 

What dollar link?

Second, history is not much help in determining what dollar weakness means for growth in Latin America.   In the post-Bretton Woods era we have seen periods in which a strengthening dollar appeared to have accompanied stronger growth in Latin America (most notably in the second half of the 1990s), but we have also seen periods where the reverse has been true (since 2002, when a weaker dollar accompanied better growth in Latin America).   Indeed, we have also seen periods such as the early 1990s when a largely flat dollar was associated with a sudden rebound in activity in Latin America. The data from the 1980s are equally mixed.

In recent decades, a strengthening dollar has been associated with rising, falling and stagnant activity in Latin America.  With that kind of track record, it is risky to come up with blanket statements regarding what dollar weakness should mean for the region.

Third, the historical relationship between movements in the dollar and movements in Latin American currencies is also inconclusive.   Whether you look at the Fed’s broad or major currency index against each currency in Latin America or against Latin American currencies as a whole, there is little clear pattern.  We’ve built a Latin American real effective exchange rate index with a fair amount of trepidation: it runs the risk of double-counting the impact of trade within the region, but it provides a useful proxy for the region’s currencies.  And it shows that in the past 25 years, the region’s currencies have gained ground with a strengthening dollar (the second half of the 1990s), gained ground while the dollar has weakened (since 2002) and gained ground when the dollar was largely stable (early 1990s). The sudden drops in Latin currencies (with and without the Mexican peso) have taken place as the dollar was gaining ground (1982), as the dollar was largely stable (1994) and as the dollar had peaked (2002).

Appreciating growth

Fourth, instead what holds up best is the association between strong currencies and strong growth in Latin America.  In each of the countries and in our imperfect Latin American real effective exchange rate index, better growth in real GDP accompanies a stronger currency.  Even after putting aside for the moment the issue of causality — does a strong currency boost domestic demand and hence GDP, or does strong GDP attract inflows, pressuring the currency to strengthen? — growth in Latin America and strong currencies have coincided for decades largely uninterrupted.

And there’s the rub: unless the dollar’s decline in the coming months begins to sharply change the growth dynamic in the region, it is hard to see a significant change in the outlook for Latin America.  Indeed, we suspect that most of Latin America is likely to benefit if we see moderate dollar weakness in 2007. 

Argentina, for example, is likely to benefit in the event of a steady, but not abrupt weakening in the US dollar. The authorities have in practice, but not officially, pegged the Argentine peso to the US dollar and not allowed it to strengthen beyond 3.05 during the past year.  But with inflation in Argentina significantly higher than that of its trading partners, the authorities have been running the risk that the peso could rapidly lose its competitiveness unless the nominal exchange rate was managed to a weaker level of say 3.1 or 3.2.  That, in turn, would likely require an even greater level of intervention by the central bank in accumulating reserves and could complicate the work of the central bank to control inflation. 

However, a weaker US dollar when compared to the major currencies should help Argentina avoid the need to push for a more devalued peso relative to the dollar.   An exchange rate of 3.05 or even 3.0 during 2007 might mean that the Argentine peso is losing some of its competitiveness relative to the dollar as Argentine inflation is running at multiples of inflation in the US; however, Argentina’s quasi-peg to the dollar in effect helps to weaken the peso relative to other currencies. 

Brazilalso should benefit as a weaker dollar helps to limit Brazil’s real effective (trade-weighted) appreciation.  But further dollar weakness is likely to put intense pressure on the widely watched BRL/USD exchange rate.  Although the central bank has always denied that its massive reserve accumulation is being driven by an attempt to limit the real’s gains, a weaker dollar substantially raises the risk that the real could break through 2.0, despite the heavy intervention on part of the central bank to buy up dollars.  With inflation running at 2.96% in the 12 months ending in March, overnight interest rates at 12.5% (after 15 cuts in nearly two years) and a healthy current account surplus near 1.3% of GDP, it is not hard to see why the real is under pressure to strengthen. 

Indeed, Mexico is likely to be the one exception in the region unlikely to benefit with a stronger currency in the event of dollar weakness. We would highlight two reasons for Mexico to stand out from the rest of the region and gain less from a weaker US dollar. First, the Mexican economy would simply benefit least in the region from a boost to commodity prices.  It is largely an exporter of manufactured goods, unlike the rest of the region.  Second, the link between Mexico and the US is strongest, meaning that any slowdown in US activity is likely to be felt most directly in Mexico and offset the gains from a more competitive currency. Indeed, the downturn in construction in US residential sector – which still seems to be having limited spillover to US consumption – appears to already be taking its toll in Mexico as remittance growth has slumped (see “Mexico: The End of Abundant Remittances”, Global Economic Forum, April 17, 2007). 

Bottom line

We certainly don’t plan on ignoring the US dollar, but our principal focus will be on global and US growth.  Unless a weaker dollar begins to seriously crimp Asian and European growth, we doubt that most of Latin America will suffer.  Indeed, while the past 25 years have produced episodes in which the region gained while the dollar rose, gained while the dollar fell and gained while the dollar was stagnant, we suspect that Latin America is likely to post good growth even if the dollar weakens from here.  Mexico might prove to be the odd one out, given its proximity to the US and the nature of its trade basket. 

 

Turkey
The Rise and (Slow) Fall of Inflation
April 24, 2007

By Serhan Cevik | London

Inflation is not coming down as fast as implied by aggressive monetary tightening. The last consumer price index reading was right in line with our estimate, and we now project the annual inflation rate to decline from 10.9% in March to 10.6% this month, hopefully signaling the start of a new disinflation phase. However, it would still be closer to the upper bound of our forecast profile, implying a year-end reading that is higher than the central bank’s target. Although we expect lower inflation readings in the coming months, the pace of correction is not as swift as the baseline scenario entailed by aggressive monetary tightening and the resulting moderation in domestic demand. After last year’s global volatility shock and the loss of credibility during the appointment of the new management team, the Central Bank of Turkey raised interest rates by 425bp to 17.5% to curb the pass-through from the lira’s weakening and limit the deterioration in inflation expectations. Even though tighter monetary conditions have already led to a correction in the exchange rate and a marked moderation in consumer spending, pricing behavior is still not consistent with underlying economic developments.

There is a disconnect between the retrenchment in consumer spending and inflation. Coupled with the uncertainty factor, the sudden and aggressive tightening of monetary conditions last summer resulted in an abrupt slowdown in domestic demand. Private consumption growth fell from 9.9% in 1H06 to 2.3% in 3Q and 0.1% in 4Q — the lowest reading since the 2001 crisis. This is a significant retrenchment, but we need to analyze the subcomponents to see the full extent of the correction in discretionary spending. For example, the consumption of durable goods — highly sensitive to interest rates and changes in credit conditions – plummeted from an annual growth rate of 14.2% in 1H06 to -8.3% in 3Q and -6.3% in 4Q. With such a far-reaching adjustment in the domestic economy, we should have witnessed the return of disinflation earlier than what the latest figures now suggest. Instead, there is a curious disconnect between the retrenchment in consumer spending and inflation dynamics. For example, despite the nose-dive in demand for durable goods, the annual rate of increase in durable good prices (excluding gold) surged from 2.8% at end-2006 to 7.2% in March.  Muted but similar behavior is also apparent in other (tradable) categories of the CPI basket, as goods price inflation climbed from 8.7% in 2006 to 10.4% last month.

Supply-side shocks explain the rise in inflation, but not the prevailing inertia. In our view, supply shocks (like higher energy, gold and unprocessed food prices) were the original culprits slowing the pace of disinflation, which has come to an end with the pass-through from the lira’s sudden depreciation last summer. Still, although the volatility of commodity prices still presents a challenge, the inflation problem has moved beyond supply shocks. For example, inflation excluding energy and unprocessed food prices increased by 330bp to 9% in 1Q07, even as the notoriously sticky inflation rate in services at last showed a marginal (10bp) improvement to 12.1%. Likewise, even if we exclude administered prices and changes in tax rates, there is still an upward trend in core inflation — from 8.9% at end-2006 to 10% in March. We think that higher import prices — stemming from the rise in inflation in countries like China — may have contributed to higher tradable goods price inflation in Turkey. For example, clothing and footwear prices already posted a 4.2%Y increase in March, up from an average of -0.2% last year. Indeed, we expect this category (and household appliances) to become the crucial determinant of overall inflation in the next couple of months. Even so, yet another factor has an overwhelming influence over domestic pricing decision, in our view.  

An ‘uncertainty premium’ in pricing decisions could explain the inflationary inertia.  The Turkish economy has become dollarized over the past four decades, as residents struggled to protect their wealth and income flow against ‘unexpected’ shocks. Although the normalization of the macroeconomic landscape led to an encouraging reverse dollarization process, global volatility shocks and election worries have disturbed residents’ portfolio allocations and pricing decisions. As a result, there is now a higher risk premium — reflecting political uncertainties and the possibility of another bout of currency depreciation — embedded in domestic prices (see Turkey: Pricing the Unexpected, February 12, 2007). Of course, this creates an intriguing bottleneck in the disinflation process and thereby forces the central bank to maintain a restrictive monetary policy stance. Although transitional challenges may become disheartening from time to time, we still believe in the secular nature of disinflation in Turkey. As long as fiscal performance remains on track, the lagged effect of tighter monetary conditions will drive inflation dynamics towards the central bank’s multi-year target. Therefore, recognizing the upside risk to our below-consensus inflation projections, we expect to see slow but steady disinflation in the remainder of the year.