Credit Suisse: Australian dollar has further to fall

From Credit Suisse:

We have just published an article (attached) explaining why we think the AUD/USD has not depreciated enough yet to generate an easing of financial conditions via the external sector. We estimate that the equilibrium level of the currency has fallen to 71c, with downside risk in the event that commodity prices fall, and yield differentials invert further.

There have been several major developments underpinning the recent depreciation of the AUD/USD. Most notable among these have been:

1. The RBA pushing back rate hikes, despite ongoing Fed tightening, causing the Australian-US nominal 10-year yield differential to fall to an historical low of -30bps recently.

2. Weakness in China and emerging market economies, causing commodity prices to fall.

But an equally significant, but less recognized development was the sharp upward revision to the US household saving rate in the 2Q national accounts by almost 4% of disposable income. This matters, because prior to the revisions, it looked as though the US trade deficit was out of synch with household behaviour. More specifically, the US trade deficit was shrinking at an unusually brisk pace, despite ongoing household dis-saving, supporting stronger US consumption of foreign goods. We had previously attributed this anomaly to preparations for Trump’s tariffs. We still believe that this is the case. However, now that the saving rate has been revised dramatically higher, it appears that the consumer has not been gearing up as heavily as we previously thought to consume foreign goods. The trade deficit has justifiably shrunk in line with longer-term fundamentals.

Going one step further, this means that we should not expect to see an imminent widening of the US trade deficit once we roll into the new tariff regime. Indeed, the trade deficit could shrink further. And if the US is a smaller net importer of goods and services, the flipside is that it is also a smaller net exporter of USDs to pay for them. This exacerbates USD shortages for the rest of the world via the trade channel. And trade matters as much as capital flows for economies like China and Australia.

In other words, we think that the USD shortage has been understated up until now, and the AUD/USD, and CNY/USD have been reflecting a false reality. Now that the benchmark revisions have come through, we should also revise our view of USD funding availability globally.

Our joint parity equilibrium framework is capable of capturing relative currency scarcity, because it captures both the demand for, and supply of currency in a single equation. As it turns out, a compromise of purchasing power, trade and interest rate parity conditions is sufficient to completely capture the longer-term drivers of pricing. Our long-run equation for AUD/USD equilibrium is:

-1 * log (relative CPI) + 1 * log (relative terms of trade) + 1 * adjusted trade balance differential + 10 * 10-year real yield differential + constant

The first term captures purchasing power parity – the law of one price. The second and third terms together capture trade parity. The third term captures uncovered interest rate parity on a 10-year horizon. The constant is the only term that requires estimation, and this is merely a normalization (indexation) exercise.

In our article, we briefly outline a novel approach to conceptualizing trade parity based on the national accounting identity and flow of funds (the “adjusted trade balance differential”). A proper understanding of this framework has profound implications for modelling the marginal availability of currency via trade, and is capable of fixing many of the known correlation breakdowns in many conventional models of currency. This makes intuitive sense – most conventional currency models only examine the demand side of the equation – but few capture the supply side, and indeed, have not had to do so until we have entered into a post-crisis regime.

The sharp upgrade to the US household saving rate implies a downgrade to the Australian-US adjusted trade balance differential, and a downgrade to the equilibrium level of the AUD/USD, because of the relative scarcity of USDs relative to AUDs via trade.

Finally, it is worth noting that our joint parity modelling approach has several benefits over conventional approaches:

1. It imposes model parameters through theory, rather than over-fits a regression. If the input variables are defined well enough, this implies that the currency model should be very stable through time, making it more reliable for use in forecasting. As it turns out, our model does pass all standard diagnostic and stability tests, with flying colours.

2. Currency deviations from joint parity equilibrium tend to mean-revert. More specifically, the actual currency tends to converge to the equilibrium level, rather than the other way around. Therefore, it is possible to profitably trade over- and under-valuation signals, even though fundamentals can shift rapidly through time.

On a mark-to-market, real-time basis, the equilibrium level of the AUD/USD is a touch below 71c. But fundamentals can, and will change. We expect that:

1. Commodity prices will come under pressure, given tight financial conditions in China and slowing growth momentum. This will put downward pressure on the Australian terms of trade, and lower exporter translation demand for AUDs.

2. Australian-US yield differentials will invert further. Weakness in China, and persistently low inflation are likely to keep the RBA on the back foot for longer, while the Fed continues to hike rates.

If commodity prices fall further by 20%, and Australian-US yield differentials invert by a further 50bps, as our leading indicators suggest, we should expect the equilibrium level of the currency to fall by at least another 12%. This would take equilibrium down to 63c.

With all of this in mind, the recent depreciation of the AUD/USD to 74c from 81c earlier in the year seems woefully inadequate to generate an easing of financial conditions for the economy via the external sector. This is because the equilibrium level of the currency is falling faster than the actual, and the premium in the currency is actually rising. Put differently, the AUD/USD is not falling for good reasons, and it is hard to arrive at a more optimistic conclusion for as long as emerging markets remain weak, and domestic activity remains soft.

Interestingly, when we read RBA statements carefully, we find that officials do not make much of recent AUD/USD depreciation. They merely state that the currency remains within its trading band of the past few years. Officials are quick to point out the costs from a stronger currency. But their view of the currency is asymmetric on the downside, precisely because the disappointing regional growth backdrop makes it hard to interpret whether depreciation is a good or bad thing.

Therefore, we do not think that recent AUD/USD depreciation is sufficient to negate the need for rate adjustments from the RBA.

From an equity market perspective, USD shortages in the emerging market complex are negative for commodities. Tight financial conditions, exacerbated by a premium in the AUD/USD relative to equilibrium, are negative for domestic cyclicals and banks. However, a weaker AUD/USD is a positive for high quality industrials with offshore earnings exposure. It is also relative positive for selected bond proxies to the extent that the currency is falling for bad reasons.