Archimedes' Dollar Smile

If you can figure out the path of the dollar then you're starting from an advantaged point in assessing where other major macro trades are headed

Author's Avatar
Sep 19, 2017
Article's Main Image

Give me a place to stand,
and a dollar thatā€™s trending,
and I can move the world
~ Archimedes

Archimedes, the Greek tinkerer, knew of the U.S. dollarā€™s importance to global macro well over 2,200 years ago, long before Fed Chair Janet Yellen was born. A man ahead of his time.

If you read my work, then you know that I refer to the dollar as the worldā€™s fulcrum.

This is no exaggeration. The U.S. dollar is that important to global markets.

The deeper you get into the global macro game the more you realize nearly every trade is a derivative of a long or short dollar position.

Long Brazilian equities? Youā€™re short the dollar.

Long airliners? Youā€™re making a macro call on oil and thus the dollar.

Long volatility? Youā€™ve got an embedded dollar call there.

Short inflation through duration? Then youā€™re long the dollar.

The dollar is important.

Why is this?

Short answer is the dollar is the worldā€™s reserve currency. Itā€™s what commodities and goods are priced in for global trade, and itā€™s the dominant global funding currency. The dollar is pervasive, and itā€™s everywhere which is why the Fed swings the biggest stick of them all.

If you can figure out the path of the dollar, then youā€™re starting from an advantaged point in assessing where the other major macro trades are headed.

Itā€™s always my starting point when analyzing any market.

Thatā€™s what Iā€™m going to spell out quickly here. Weā€™re going to run through some basic models for thinking about the dollar and then weā€™ll jump into the bearish thesis, the bullish thesis and conclude with where in the argument I sit.

And just to make clear my current biases. Our team at Macro Ops has been short the dollar against the Australian and Canadian dollars since June 27 ā€“Ć‚ until this last week when I closed out for profit.

I took these trades with the belief that they were countertrend moves. I was in the cyclical dollar bull camp but was bearish over the intermediate term, with the expectations of a 10% pullback due to technical, sentiment and macro reasons.

Now that the move has played out, I need to update my view.

Some dollar models

Iā€™ve shared a more in-depth piece on the way I think about FX that you can find here. In it, I hash out George Sorosā€™ arrows and the core/periphery model.

Today weā€™re going to talk about the ā€œdollar smileā€ concept put forth by Stephen Jen of Morgan Stanley because itā€™s relevant to our conversation.

The theory is simple. It states that the dollar tends to outperform when the U.S. economy is very strong (on the left side of the smile) or very weak (right side). And it does poorly when the economy is just muddling through (middle of the smile).

19Sep20171138591505839139.jpg

Why is this?

Well the logic is straightforward.

The U.S. trades at a ā€œsafety premiumā€ relative to other countries.

Some might snarl at that, and there are a lot of gold bugs who think the U.S. governmentā€™s debt blowup is imminent. But the reality is that relative to the rest of the world, the U.S. has one of the most dynamic economies, highly liquid markets, (relatively) stable governments, decent rule of law, and again, weā€™re the worldā€™s reserve currency.

When the U.S. economy is performing well, investing in the U.S. is a no-brainer. And since well over 80% of the moves in the FX markets are due to speculative flows, this fact matters.

When the U.S. economy is very weak, the dollar performs well because it gets a safety bid. Money gets pulled back from overseas to within safer borders.

Most international funding is done in U.S. dollars. And when volatility increases and markets are perceived as riskier, these dollar loans are called back and Brazilian reals, or whichever, get converted into dollars to cover the dollar debt thus putting upward pressure on the dollar.

But when the economy is in the middle of the ā€œsmileā€ and just muddling through, the dollar tends to perform poorly. Why?

Well, the primary reason is that mediocre growth and low inflation is bullish for risk assets because it keeps the Fed steady and prevents it from raising rates too quickly.

So a steady Fed keeps interest rates low. These low interest rates suppress volatility and push investors further out the risk curve in search of returns. They go overseas to high growth emerging markets to play in their fertile fields.

Low growth and low rates lead to speculative outflows from the U.S. which drive the dollar down relative to where the capital is flowing.

Thereā€™s a reflexive relationship in these FX flows that starts to dominate.

This is because currencies make up the largest piece of the total return picture when investors put money into foreign markets.

So when speculative capital leaves the U.S. because of low rates and slow growth and then flows into, say, Latin America, it depreciates the dollar while appreciating the Latin American currencies. This drives up the total return of those investments in these EMs which then attracts more speculative flows (i.e., reflexivity).

Hereā€™s a short snippet from George Soros on the mechanism.

To the extent that exchange rates are dominated by speculative capital transfers, they are purely reflexive: expectations relate to expectations and the prevailing bias can validate itself almost indefinitely. Reflexive processes tend to follow a certain pattern. In the early stages, the trend has to be self-reinforcing, otherwise the process aborts. As the trend extends, it becomes increasingly vulnerable because the fundamentals such as trade and interest payments move against the trend, in accordance with the precepts of classical analysis, and the trend becomes increasingly dependent on the prevailing bias. Eventually a turning point is reached and, in a full-fledged sequence, a self-reinforcing process starts operating in the opposite direction.

This is why currencies tend to trend for long periods of time once they get going. See the seven-year cycle below.

19Sep20171138591505839139.jpg

Strong U.S. growth equals outperforming dollar. A very weak U.S. economy equals strong U.S. dollar. And a muddling U.S. economy equals weak dollar.

I should point out an important point that Iā€™m not sure Stephen Jen mentioned when he introduced this concept. But all of this is relative. We donā€™t care about the U.S.ā€™s economic growth on an absolute basis. We care about its economic picture relative to that of the rest of the worldā€™s (ROW).

If the U.S. is muddling at sub-2% growth but emerging markets are a dumpster fire, like they were following the GFC just up until this last year, then thatā€™s still dollar bullish.

The idea is that investors have to perceive risk to be low and the reward to be well above the U.S.ā€™s premium in order for capital to leave our beautiful shores and invest in a Chilean poultry producer.

The chart below shows this dollar smile relationship at work. Blue line shows the U.S.ā€™s growth relative to the rest of the world's (ROW) and orange line is the dollar.

19Sep20171138591505839139.jpg

It works because if growth in the U.S. is strong relative to the ROWā€™s, then it means that the Fed is likely leading developed markets in raising interest rates. This makes the U.S.'s real interest rate spread more attractive. And vice versa when the U.S. is performing well below the ROW.

And as Soros said, the ā€œexpectations relate to expectationsā€ that drive FX trends far from their ā€œfundamental equilibriums.ā€ This is why we always need to be trying to understand the markets expectations around future relative growth and rates.

Once we understand the embedded expectations we can compare them to the likely outcomes and see if thereā€™s a divergence between the two.

If there is, then we can identify potential catalysts that would reprice these expectations to be more in line with reality. And then we have a trade.

The dollar bull market that began in 2011 was driven by strong relative U.S. growth (seen in the chart above) and the Fed signaling tighter monetary policy relative to its developed market peers.

This created a reflexive loop that kicked off in 2014. Stronger growth and signaling of tighter policy drove the dollar higher along with U.S. equities. This brought in further speculative flows and kicked that reflexive process into gear.

19Sep20171139001505839140.jpg

But by 2016 market expectations for U.S. outperformance and tighter relative monetary policy began to exceed likely outcomes.

Long dollar positioning had become crowded, and the trend extended.

Europe and emerging markets on the other hand were recovering from very low bases and horrible sentiment as well as benefiting from improving global economic growth.

This set up a reversion in expectations.

The market began repricing the Fedā€™s rate path lower while pricing in tighter monetary policies for other DM central banks.

This caused the huge unwind in the dollar, which is currently the greenbackā€™s worst year on record (chart via Bespoke).

19Sep20171139001505839140.jpg

The market is now overweight the eurozone and emerging markets while underweight the U.S.

19Sep20171139011505839141.jpg

And expectations are high for Europe and EMs while neutral for the U.S.

19Sep20171139011505839141.jpg

There have also been signs that DM central banks have been coordinating policy and using forward guidance to move exchange rates, letting FX markets do the heavy lifting for them in their monetary policy goals.

Learning from 14ā€™-15,ā€™ where diverging monetary policy drove the dollar higher and risked pushing the global economy over the edge, central bankers started managing expectations in order to subdue the dollar while the Fed moved to become the first to begin reducing its balance sheet.

I wrote about this signaling by the Game Masters (the central banks) in a brief that went out July 24. For reference, Iā€™ve included the excerpt below.

Basically, Brainard stated how a country with a large existing trade deficit (i.e., the U.S.) should choose to tighten its monetary policy in a way that puts the least amount of pressure on its currency (the dollar) and its exporting sector.

The opposite is true for a central bank of a large surplus region (i.e., Germany/ECB), which should tighten in a way that helps bring its economy into better external balance.

Lael argues that tightening by increasing the policy rate has a bigger impact on the exchange rate than tightening through balance sheet reduction. Therefore, the U.S. should tighten through balance sheet reduction while the ECB should tighten through interest rates. And if the two coordinate, they can reduce the potential for instability (i.e., large China yuan deval or a blowout in Italian yields).

Now whatā€™s currently happening in the land of central banks?

The Fed has recently come out as dovish on hiking but signaled it intends to start letting its balance sheet run off. And other major central banks, like the ECB, have come out surprisingly hawkish, indicating that the beginning of rate normalization is upon us.

This is a large reason the dollar has sold off. Most countries are better off with a slightly weaker dollar and tighter local interest rates than they are with ever-diverging rate paths and a strengthening dollar leading to tighter global liquidity.

Everybody is focusing on interest rate spreads and dollar debt ā€“ which do matter, just less so at this time ā€“ while failing to factor in the new intentions of central bank coordinated policy. The CBs donā€™t want a repeat of what a stronger dollar did to global markets in 2015. Theyā€™ve wised up (a little bit at least).

The price action of DXY definitely seems to be confirming this, finishing near its lows for the week and breaking through a significant line of support. The chart of DXY is on a weekly basis and it falls down to at least its 200-week moving average (the blue line) and likely even a bit further before finding a significant bid.

The last nine months weā€™ve been in the middle of the dollar smile curve. The U.S. economy has been muddling through while the ROW has rebounded off its lows and DM central banks have played catch up to the Fed.

This is what the bear case for the U.S. dollar is based on, that weā€™ll continue to stay in the middle of this curve.

Hereā€™s a summation of the thesis from Nomura.

19Sep20171139011505839141.jpg

And Mark Dow also summed up the case well, writing:

On balance, a weaker dollar. The U.S. shifting to the balance sheet from rates as the front-burner tool, and global CBs ā€˜catching upā€™ to the Fed are the major drivers. Euro, Australian dollar, EM all good. The positioning/psychology in the dollar doesnā€™t strike me as too extreme, so dollar weakness doesnā€™t have to be dramatic.

The question that Iā€™m most interested in when looking at the dollar now is: How long is this ROW outperformance and DM central bank catch up likely to last relative to what is already priced into markets?

19Sep20171139021505839142.jpg

And itā€™s here where I have trouble buying fully into the continuation of the dollar bear thesis.

When looking at credit spreads and EM debt, you can see that risk is increasingly priced out.

That means that the embedded expectations are high.

This tells me thereā€™s an exceedingly narrow path of outcomes that the future needs to meet for this trade to continue to work.

Second, and maybe even more important, it appears the low-growth, low-inflation narrative has been fully adopted, not just by the market but also by the Game Masters themselves.

Fed President Bill Dudley said this in a speech recently:

While some of this yearā€™s shortfall can be explained by one-off factors, such as the sharp fall in prices for cellular phone service, its persistence suggests that more fundamental structural changes may also be playing a role. These include the increased ability of prospective buyers to compare prices across different sellers quickly and easily, the shift in retail sales to online channels of distribution from traditional brick-and-mortar stores, and the consequences of these changes on brand loyalty and business pricing power.

A finance friend of mine recently attended an investors conference where the main subject was inflation. And he mentioned that this was the first time, in the many years heā€™s attended, when the crowd wasnā€™t expecting higher inflation.

19Sep20171139021505839142.jpg

Instead, everyone was talking about how the Phillips curve is dead and how technology has brought on a new secular trend of permanent low inflation.

This popular narrative is priced into all areas of the market. And it has been somewhat true up until recently.

But I chuckled when I heard this because for the first time in a long time Iā€™m seeing data that says inflationary pressures are building and higher prices are right around the corner. I covered this in our most recentMIR.

The market is pricing in a much lower hiking schedule than the Fedā€™s already shallow projected path (see chart below). Fed fund futures arenā€™t pricing another rate hike until June of next year.

19Sep20171139031505839143.jpg

With inflationary pressures building, it seems to me that thereā€™s a large possibility that the market is on the wrong side of the inflation trade ā€“ and therefore, is on the wrong side of the dollar and EM trade, as well.

And if we do enter an inflationary environment, do you think Europe will be able to stomach a repricing of rates higher, better than the U.S.?

Count me as skeptical.

19Sep20171139031505839143.jpg

There are a few outcomes where the dollar bear case becomes more likely, and theyā€™re centered around the Fed.

With the recent and unexpected departure of Fed vice chairman Stanley Fischer, the Trump administration now has five Fed seats it can fill should it choose to replace Janet Yellen.

Thereā€™s a lot of speculation that Trump will install loyal puppets, and weā€™ll return to a dollar-crushing era similar to that of the Nixon-Burns partnership of the '70s.

This is possible, and I donā€™t have an edge in predicting who the president will choose, so Iā€™ll have to react to that when the time comes.

But I do have a hunch that weā€™ve hit bottom in the Trump presidency and that things are likely to improve from here (at least somewhat).

With the divisive influencers out of his Cabinet and former Marine Gen. John Kelly now effectively steering the ship, it seems to me that weā€™re seeing and hearing less from the commander in chief, especially on Twitter, which is a good thing.

And with all likelihood of positive economic policies such as tax cuts and infrastructure spending priced out of the market, it seems like an opportune time for some positive surprises to the upside in the U.S.

Plus, the recent devastation from Hurricanes Harvey and Irma are likely to quell infighting within Congress, at least for a little while.

The U.S. appears ripe for a period of positive expectation revisions.

19Sep20171139031505839143.jpg

The way I see things is that, for the dollar bear case to continue to play out, a large number of things have to unfold to match the expectations that are priced into markets.

  • Volatility needs to remain low.
  • Investors need to remain in risk-on mode.
  • Inflation needs to remain low.
  • The Fedā€™s hike path needs to come down to meet the marketā€™s expectations.
  • DM central banks need to continue to signal tighter policy ahead.
  • The U.S. needs to continue to muddle through while the ROW goes on a tear.

All the dollar bull case needs is just for anyone of these things to not happen.

The dollar bear case needs to walk a shaky tightrope to continue to play out while the dollar bullish case can spawn from any number of outcomes.

In addition, when you throw in the Fedā€™s likely coming balance sheet reduction (likely to start this month) and the Treasuryā€™s cash normalization ā€“ which due to the debt ceiling debate being moved to December, now wonā€™t be happening until early next year ā€“ you have some large liquidity drains that will begin to open in the coming quarters.

Draining liquidity leads to the repricing of risk. And with risk priced out of current markets (especially credit markets) thereā€™s the likelihood that this repricing will become jet fuel for the dollar.

The good folks at Nordea Markets have covered this more in depth, and you can read about it here.

The longer dated euro/dollar basis swaps are beginning to price this in while the near market is not.

19Sep20171139041505839144.jpg

As of right now, I put the odds in favor of the dollar bull thesis.

I donā€™t know when this dollar selloff will end. Though positioning and bearish sentiment are reaching extreme levels.

19Sep20171139041505839144.jpg

Iā€™ll be patient and wait for higher inflation numbers, then see how the Fed responds.

Iā€™d also like to see sentiment become more negative on the dollar than it is, but Iā€™m not sure weā€™ll get that.

I continue to use the late '90s analog for the macro environment and have found it extremely helpful to understanding the possible outcomes for today. Itā€™s far from perfect but helpful, nonetheless.

So, of course, strong convictions, weakly held.

The trading game is not about forecasting, being right, or showing everybody how smart you are. Itā€™s about making money.

Iā€™ll be quick to rerate my probabilities if new disconfirming evidence comes in, such as inflation falling further or Trump replacing Yellen with Alex Jones.

Until then, pay attention to the dollar smile and look for comparative growth relative to expectations.