An Anatomy Of Long-Term US Dollar Cycles

Authored by Viraj Patel, FX & Global Macro Strategist at Arkera, originally posted by the author’s Medium page.

Over the last 45 years, a typical US dollar trend cycle has lasted around 6 to 7 years — with the Broad Real Dollar index posting an average change of 34% per cycle. Assuming that we’re in the midst of a dollar bull run that started in August 2011 — the current upswing to date has posted a 36% rally and is almost 9 years in duration. So, is the dollar about to peak and embark on a multi-year bear trend?

In short, no — at least not yet. When we look at the cyclical and structural macro conditions that have underpinned major downswings in the dollar since 1975, none of these are currently being met. Instead, the backdrop that is set to prevail in the initial stages of the COVID-19 global economic depression is likely to remain supportive for dollar haven flows.

Characteristics of a Dollar Bear Trend

Taking a deeper look into the factors that have underpinned major cyclical moves in the dollar — we see that bear trends typically show the following 6 common characteristics:

1. A substantial decline in the USD’s relative interest rate advantage — with the dollar always exhibiting negative carry
2. Stronger GDP growth in the Rest of the World (RoW) relative to the US — thus declining US-RoW growth differentials
3. Rising US twin deficits driven by a joint widening of fiscal and current account deficits (bear cycles start 2–3 years after initial widening)
4. High global trade and manufacturing activity (periods of above-trend world trade growth)
5. Money being put to work in RoW assets — driven by relative underperformance of US equities and low market volatility
6. Official US Treasury FX intervention (prior to 2000) — coordinated across major countries with the aim of achieving a weaker dollar

Aside from the Fed’s balance sheet, there’s very little working against the dollar at present. There is too much uncertainty to confidently say whether the RoW will be able to recover from the Global Coronavirus Pandemic materially faster than the US — whilst global trade uncertainty driven by a protectionist US administration is likely to continue weighing on cross-border activity and RoW asset performance for the foreseeable future. The dollar is normally the best place to hide in the FX world when there is this level of uncertainty around in global markets.

If anything, the relationship between the Fed’s balance sheet and the dollar mirrors the 2008/2009 crisis playbook — where the USD continued to strengthen even after the Fed’s initial QE intervention. The broader macro backdrop of a global recession, weak cross-border trade and relative underperformance of RoW assets also lends support to the idea that the dollar will behave in 2020 just like it did in 2008/2009. The main difference, however, is that back then the dollar entered the crisis from a much fundamentally weaker standpoint — the BIS USD REER index was around 15% below its 10-year average in July 2008. In February 2020, the dollar was just over 14% overvalued based on the same measure. This, coupled with the Fed’s relatively more aggressive response, may prove to be a strong limiting factor for how much the dollar can strengthen in the current crisis relative to 2008/2009.

What is needed to set in motion a multi-year dollar bear trend?

If the 6 common characteristics of a USD bear trend are anything to go by — then the next few years would need to see a distinctly different cyclical and structural backdrop for the dollar if it is to experience a 20–30% decline in line with historical bear trends. This would involve: (1) deeply negative US rates and persistently weak US GDP growth relative to the RoW; (2) a wider US current account deficit and above-trend global trade growth; and (3) underperformance of US equities and low market volatility. We explore the likelihood of each of these scenarios unfolding below.

1. Deeply negative US rates and relatively weaker US GDP needed to see a weaker dollar

Whilst the Fed’s aggressive easing actions so far have eliminated the USD’s carry advantage — it may not be enough to see a weaker dollar.

US relative rate differentials against major trading partners have declined around 180bps since Oct 2018. But it’s still not enough to see negative USD carry given the number of developed countries either at the Zero Lower Bound (ZLB) or with Negative Interest Rate Policy (NIRP). The Fed has an impossible task of driving US rates low enough to materially impact the dollar — especially when we’re in midst of a race to the bottom across global central banks. As we’ve seen in New Zealand last week, most developed market central banks are not done easing yet — we should certainly expect more from the ECB, BoE and BoJ in the coming months. On the flip side, we think the Fed will be hesitant to take policy rates below 0% given the market structure in the US and the negative implications for banks and money market funds — meaning we’ve reached a point where US rate differentials (even when adjusted for relative central bank balance sheet expansion) have likely bottomed out for now. If that’s the case, the 180bps adjustment in rate differentials in the last 18 months is not sufficient to trigger a USD bear cycle — the past two major dollar downswings have seen a downward adjustment in relative US rate differentials by around 300bps.

Look for ‘Fed last to hike’ or Yield Curve Control for a weaker dollar heading into 2021. Policy divergences may become more apparent as we transition from the global downturn to the global rebound phase — with economies exhibiting different recovery paths based on domestic factors (COVID-19 second wave, ability to successfully reopen economies, varying hysteresis effects). As short-dated global yields begin to recover, one way US rates could be anchored lower for longer is if the Fed adopts some form of Yield Curve Control (or flexible QE) — that sees the pace of asset-buying marginally increased to curb any pre-emptive shift higher in the US yield curve. Certainly, if any ‘Fed last to hike’ sentiment were reflected in market pricing — then short-term US rate differentials could move lower in a way that would see a weaker USD. But given where we are today — and still grappling with the fallout from the crisis — we may be a good 6–12 months away from even considering this scenario from being priced into markets.

On the growth side, the US economy is not currently expected to be hit worse than the RoW from the Corona Crisis — although it really is too early to call. Our Dollar Smile Theory work shows that global economic recessions yield a strong dollar environment — and this should be the baseline for how the dollar performs in 2020. Beyond this, the US economy would have to materially underperform the RoW for the dollar to weaken over the next few years — and it may require something like a US health policy mistake (opening up the economy too fast) for this divergence to occur. RoW-US growth differentials tend to widen by around 2ppts during USD bear cycles. Even if the US recovers slower than other major economies (not the consensus scenario), it is highly unlikely that we’ll any significant divergence in the recovery paths that would warrant a substantially weaker US dollar.

2. A wider US current account deficit and above-trend global trade growth are necessary for a dollar bear trend

Unlike in 2010, ongoing US-China trade uncertainty puts a dampener on the prospects of a big post-crisis global trade rebound

A bigger twin deficit in the US is almost certain for the next few years — not least given the large fiscal stimulus that has been deployed by Congress to fight the US economic lockdown. The US government primary deficit is expected to reach 13.5% of GDP in 2020 (versus 3.6% in 2019) — on par with what was seen during the Great Financial Crisis. However, there is a high degree of uncertainty over how the current account deficit evolves in the coming years given the current US administration’s protectionist policy stance.

Global trade growth has already been on a declining path since 2019 following the onset of US tariffs on Chinese imports. With US-China trade and geopolitical uncertainty set to continue in the run-up to the November 2020 US Presidential Elections — one has to question the ability and extent to which global trade can rebound in 2H20. Indeed, the minor USD bear trend that followed the GFC between March 2009 and July 2011 was underpinned by several years of above-trend global trade growth. Given the circumstances, it is highly likely that this time may be different.

3. Global investors need to put money to work in cheap RoW assets for the dollar to weaken

US equities continue to outperform in 2020 despite their relatively expensive valuations — suggesting that global investors still see US stocks and credit as the best of a bad bunch

The recent USD bear trends of 2002–2008 (-25%) and 2009–2011 (-17%) were both characterised by greater global portfolio shifts towards RoW assets — with US equities underperforming during these periods. Of late, however, there has been a clear preference by global investors for US equities — with the S&P 500 posting an average annual outperformance of 9.8% since August 2011 versus its global peers. This relative outperformance has only accelerated since the Corona Crisis hit global markets (albeit stock markets across the world have been very choppy). There’s little to suggest that this trend will turn on its head anytime soon — much like the US relative growth story, there would need to be an idiosyncratic US shock from the current crisis that clearly impedes the US recovery relative to the RoW. With the Fed also making sure this doesn’t happen by buying just about anything it can get its hands on — it is highly unlikely that US policy inertia will be such an idiosyncratic shock. Instead, one would have to look to arguments such as a greater risk of a second COVID-19 wave in the US or a bigger US corporate default cycle to justify US equities from materially underperforming the RoW from here. It’s too early for anyone to make this their central scenario right now.

And what about the possibility of official US Treasury FX intervention to weaken the dollar…?

The Broad Real Dollar Index has never embarked on a multi-year bull run when it has been more than 7% overvalued in the last 50 years. That’s because the 1980s and 1990s were defined by large-scale coordinated FX interventions across major economies — aimed at either weakening or strengthening the dollar whenever it reached extreme levels of misvaluation.

The idea of official US Treasury FX intervention to weaken the dollar has gained greater airtime in recent years — not least due to anti-strong dollar comments by President Trump. This debate certainly needs its own separate article (see here for a quick primer). But here’s the bottom line. Is it possible for the US to unilaterally intervene in FX markets to weaken the dollar? Yes. Will they? Highly doubt it. As the experience over the 1980s and 1990s shows, official FX interventions by authorities tend to do more harm than good. There are solid reasons why the dollar is strong right now — not least portfolio-related flows. Governments attempting to correct any arbitrary concept of misvaluation would only create more uncertainty in an already highly uncertain investment environment. But another 10% rally in the dollar — and, well, the idea of unilateral US FX intervention stands a greater chance of becoming a reality.