G10 FX short-term valuation: Most currencies have further to run vs USD

We used our short-term fair value model to gauge any mis-valuation of major G10 FX pairs. The model includes the following factors: a) 2-year swap rate differentials; b) relative equity performance; c) relative shape of the yield curve; d) the gauge of global risk sentiment; and e) for commodity currencies also a measures of various commodity prices. In the model, we use rolling betas over a fixed window to ensure the fair values adjust to the changing dynamics and the importance of the drivers behind the currency.

Figure 1 below provides a snapshot of the current mis-valuation in the major G10 pairs, differentiated by USD and EUR crosses.

Among the USD-crosses, EUR, JPY, AUD and NZD all show a level of undervaluation against the USD of at least 1.00%. A useful measure of comparison is the 1.5 standard-deviation band (grey bars in the chart), which indicates whether the level of the mis-valuation is stretched or not. Given that all currencies find themselves within the 1.5 standard deviation band, there is no sign of a meaningful / extreme mis-match vs their short-term fair values. But as Fig 1 shows, among the dollar crosses the EUR, JPY and NZD are closer to the bottom of their 1.5 s.d. lower bound, compared to CAD and AUD.

Among the euro-crosses, sterling is approximately 1% overvalued. This does not come as a huge surprise given the GBP gains of past weeks. It is also important to note that GBP still remains comfortably within its 1.5 s.d. band. While GBP shows a sign of modest overvaluation, the other European currencies (SEK, NOK and CHF) are much closer to their short-term fair values.

As shown in Figure 2 below, USD/JPY is by far the FX pair with the highest positive correlation to the UST 10Y yield. With the latter having moved from the 0.90% area at the start of the year to 1.25% at the time of writing, it is no surprise that the yen has been the worst performing currency in 2021.

Our short-term fair value model does account for government yields, so that a rise in US yields pushes both the fair value and the spot of USD/JPY higher. The fact that – as shown in Figure 1 – JPY presents the highest undervaluation to USD indicates that the rise in USD/JPY spot has been wider than what our fair value model would have implied. In other words, markets have added a risk premium to the pair. This may be related to expectations that US yields will continue to rise, or to mounting speculation that the Bank of Japan will cut rates at the next policy meeting (18-19 March).

JPY undervaluation to USD is close to the bottom of its 12-month 1.5 standard-deviation band, which could suggest a more limited scope for further rallies in USD/JPY.

EUR/USD is currently trading in the undervalued territory, after showing signs of overvaluation earlier in the year, when it reached the local maximum of 1.2327. One reason behind the current EUR/USD undervaluation is in our view the speculative positioning adjustment, with speculators in part cutting the prior meaningful EUR/USD longs in late January – triggered by the mix of the slow eurozone vaccination process (Figure 3) and the fiscal stimulus related re-rating of the US growth outlook – thereby driving the pair below its fair value. But with the bulk of the USD consolidation likely behind us and the theme of reflation still having more legs to go, we expect the EUR/USD undervaluation gap to fully close and EUR/USD short-term fair value to move higher over the coming months.

With the risk of hard Brexit out of the way and the fast UK vaccination process in place, the GBP trading dynamics should shift and the currency is now set to experience more frequent occasions of an overshot into the overvalued territory vs the prior Brexit-related undershoot into the undervalued territory (the latter is evident in Figure 4).

However, given the relatively fast correction in GBP higher over past weeks (sterling is the second best performing G10 currency this year) and the fact that EUR/GBP currently trades in the undervalued territory (albeit still within the 1.5 s.d. band) this suggests that the pace of GBP appreciation should slow from here.

NZD/USD presents a 1.2% undervaluation, according to our model’s estimates. The Kiwi $ has likely faced a fierce squeezing of its long positions since mid-January, with CFTC data showing a drop in net speculative positioning from +35% to +23% of open interest from 19 January to 2 February. However, we do not see the (albeit moderate) risk premium on NZD as justified, especially when compared to AUD (which presents a similar mis-valuation).

Before the Auckland lockdown was announced over the weekend, New Zealand was a virtually virus-free and restriction-free country. But, as shown in Figure 5, the build-up of NZD risk premium started already in January, so it is not a consequence of the Auckland lockdown. AUD, instead, had been facing a relatively more complicated contagion situation (still, not comparable to European or US standards), as Perth first and the State of Victoria then were put into lockdown: this may have warranted a build-up of some AUD risk-premium.

On the economic side, New Zealand is the only major economy that has moved back above pre-pandemic activity levels, and is facing limited geopolitical risk compared to Australia, where the unresolved trade-spat with China would also justify a deviation from its short-term fair value.

NZD’s undervaluation may be a consequence of expectations around a more bearish stance by the Reserve Bank of New Zealand at next week’s meeting, but we think that – considering the improved economic outlook and resilient inflation – the RBNZ will hardly get any closer to the Reserve Bank of Australia’s dovishness.

All in all, as the NZD undervaluation looks less warranted than the AUD undervaluation (which could be even higher considering the downside risks from the Aussie-China spat), and with economic and monetary divergence also constituting a bearish argument, the balance of risks for AUD/NZD appears tilted to the downside in the short term. We expect the pair to move back below 1.06 in the next weeks.