Amid perennial gyrations in global financial markets, fears of a post-QE world, and schizophrenia over EMs, risk correlations have varied in the post-Lehman era with some notable constants. For example, the risk premium in EM FX is now trading back in line with equities rather than G10 FX, presenting new trading opportunities.
So far this year, even during the mini blow-up in EMs, intra-EM FX correlations are quite low, says Jens Nordvig, managing director of currency research at Nomura Securities International, in New York. This signals that investors are still, to some degree, differentiating between the individual currencies – a healthy sign relative to indiscriminate selling, such as that observed in June 2013.
 |
Additionally, it appears that cross-market correlations, such as those between FX moves and local rates moves, are still fairly low in most EMs, and, although they are somewhat elevated in Brazil, India, Russia, Turkey and South Africa, notably, even in these markets they are not extreme.
“This is important, [because] if FX moves occur independently of local rates, it is possible to actually generate easing of financial conditions through FX depreciation, which will be supportive for growth [although it may delay current-account adjustments],” says Nordvig.
Equally striking, he adds, is that EM FX moves have no longer underperformed risk assets, with the weakness observed in EM assets having instead been consistent with moves in global risk assets – from S&P 500 to USDJPY – rather than showing any clear evidence of underperformance and EM-specific tension.
 |
In this respect, some countries appear to be normalizing monetary policy, after a few years of extraordinarily low rates – with key examples being Turkey and Brazil, and to some degree Russia and South Africa – which should stabilize FX moves over time.
“The composition of external liabilities is such that the risk of negative balance-sheet effects is smaller – that is, the foreign-currency proportion of borrowing is low – and the cumulative adjustment in currency rates is starting to add up, with the ZAR, TRY, INR, IDR all having depreciated 15% to 20% against the USD over the past year, suggesting that value is being created,” says Nordvig.
 |
The upshot of these changing correlations, he concludes, might still be to maintain a net long USD/EM position over the coming couple of months, but at the same time to take advantage of elevated implied volatility, which looks too high for a non-crisis adjustment by selling dollar calls against EM.
 |
As an adjunct to the EM picture, particularly to the rate of China’s economic growth, are the perceived risks to a country’s equity, bond and FX assets, based on the relative correlations between it and commodities prices, says Marc Chandler, global head of currency research for Brown Brothers Harriman, in New York.
The following chart breaks down the shifting sands in risk correlations.
 |
In this context, says Aleksandar Timcenko, global macro strategist for Goldman Sachs, in New York, the equity markets of Russia, South Africa, Canada, Australia and Brazil have outsized positive correlations to oil price risk, whilst at the other end of the spectrum, only Switzerland has outright negative correlations to oil price shocks.
 |
Unsurprisingly, he adds, the trade-weighted currencies that strengthen the most when the oil price rises are the Canadian dollar, Australian dollar and Brazilian real, joined by South Africa and Indonesia, whilst the currencies that weaken the most are the US dollar, Chinese yuan, Indian rupee and Japanese yen.
 |
“In trying to better understand these results, we find that net oil exports as a share of GDP are fairly correlated with estimated equity market oil leverages [correlations], although currency betas are uncorrelated with an economy’s oil sector,” says Timcenko. “But we also note that the regression results for both currencies and equity indices tend to identify a similar set of high-beta and low-beta countries.”
 |
In this context, Australia, Canada, South Africa and Brazil are at the high-correlation end of the spectrum in both markets, with the US, Switzerland and perhaps Japan at the low/negative-leverage end of the spectrum.
For G10 currencies, monetary policy expectations are the strongest driver, while carry has an inverse correlation to FX, meaning the change rather than level of interest rates has become relatively more important. This benefits GBP and SEK at the expense of CAD and AUD, and relative growth trends – as shown by the three-month change in benchmark PMIs – are also important, says George Saravelos, FX strategist for Deutsche Bank, in London.
 |