Right now, in the wake of the French 1st round election results, the Fed futures market is saying there’s a 72% chance of a Fed rate hike by June. So the obvious question is what this means for the economy, especially given recent yield curve flattening. I am cautiously optimistic. But I have some caveats and list a few places to look for signs of weakness below.

First, let’s acknowledge that for the past several years, it’s seemed almost as if investors have been waiting for the next financial crisis, fearful that the last one presaged yet more to come. But that ‘next’ crisis hasn’t happened. Instead, markets have gone from strength to strength – equity market in particular.

But it’s the credit markets where the rubber hits the road. After all, it was credit writedowns of mortgage-backed securities which ushered in the last financial crisis. Since then, credit markets have passed each and every test with flying colours. And the meltdown in the energy credit markets in 2015 and 2016 were a very big test indeed.

But of course, we haven’t had to contend with a Fed rate hike chain. And now that the Fed is finally normalizing rates, this will test markets because the Fed has signalled that it intends to continue to raise interest rates despite some signs of weakness in economic and credit data.

To date, the Fed’s policy normalization efforts have led to a lot of curve flattening. For example, the 2-10 year Treasury spread has gone from 2.66% three years ago when 2014 began to as low as 77 basis points in August. Right now, the differential is just barely above 1%.

And this flattening has all occurred as a result of tapering asset purchases and three 25 basis point rate hikes.

Let me say here that I think yield curve flattening in a Fed hiking cycle is not an a priori sign of economic weakness, though it is usually interpreted as such. What I am going to say on this is a bit wonky but bear with me because I think it matters.

What we are seeing here is not just about potential weakness reflected in a flat curve suggesting an anticipated unwinding of Fed hikes down the line. It also a reflection of market liquidity problems. I’m talking about rehypothecation here. Now, hypothecation is the term used for lending against underlying collateral. And this is par for the course in the global financial markets. But then there’s rehypothecation – which is the RE-lending of that underlying collateral by the financial institution to which that collateral was lent. As Niels Jensen described it here at Credit Writedowns 5 years ago, “Rehypothecation is a somewhat more complex lending practice and presents a much bigger risk to financial stability. It typically occurs in the world of prime brokerage which is the department within banks that services hedge funds and other customers who wish to use leverage in their portfolios and/or short stocks.”

Think of it as shadow bank money. It’s not real money that gets created but credit based on leveraging of safe assets like Treasuries, shadow bank money if you will. The key, though, is that these ‘money’ deals in the shadow banks are derivative. They’re not really money. Daniela Gabor and Jakob Vestergaard explained this last April:

A promise backed by tradable collateral remains acceptable as long as lenders trust that collateral can be converted into settlement money at the agreed exchange rate. The need for liquidity may become systemic once collateral falls in market value, as repo issuers must provide additional collateral or cash to maintain at par. If forced to sell assets, collateral prices sink lower, creating a liquidity spiral. Converting shadow money is akin to climbing a ladder that is gradually sinking: The faster one climbs, the more it sinks.

In the wake of the financial crisis, markets are much more circumspect about the quality of safe assets they use as collateral. So-called ‘private-label safe assets’ are out. European and US  private sector securitization declined from over $3 trillion in 2007 before the crisis to less than $750 billion in 2010.

Why this matters. The hunger for safe assets is immense. Private-label assets are not as well-regarded after a major blowup in this asset class. At the same time, people are starved for yield. And governments are trying to reduce deficits, necessarily reducing the supply. Irrespective of what the Fed does, there is downward pressure on long-term rates simply because the demand for safe assets is so high.

So what happens if the Fed hikes another three times this year?

Here are three places to look for clues on how the market will respond to Fed hikes. If these areas do well, the flattening is nothing to fear and will reverse itself. But if these areas do poorly, the flattening will continue and higher risk credit markets will be vulnerable.

First, in terms of the real economy, a good real-time gauge on how jobs and income should impact consumption is jobless claims. This is a series that has always been ticking up before every recession. And right now there has been no rise in the number of people losing their jobs and filing for unemployment insurance. In fact, it’s just the opposite; claims were at record lows last year, but have fallen even lower this year. Right now, the number is consistent with an economy adding around 200,000 jobs per month. However, if this number starts to rise, it is a real-time indicator of a weakening jobs picture that will feed through negatively into income and consumption.

Second, on the consumer side, as the most buoyant consumer credit product this cycle, auto loans are the canary in the coal mine. The auto subprime delinquency rate is at the highest level since the financial crisis. Auto discounting is increasing as are inventories. And auto sector financial companies like Ally have already begun to warn investors on earnings. Of course, we’ve seen this picture before in 2014 and the sector righted itself, even after the energy sector turned down. Nevertheless, while auto lending ebbs and flows do not coincide neatly with the business cycle, if weakness there coincides with weakness in other credit arenas like commercial real estate, the writedowns could be significant enough to cause a broader pullback in credit. One should also see the auto sector as a bellwether of health in the consumer credit sector.

Third, when we look at how businesses are responding, commercial real estate is perhaps more important than capital investment because of how the credit dynamics could impact business spending and financial markets. The Office of the Comptroller of the Currency’s most recent Semi-annual Risk Perspective from last fall said “continued incremental easing in underwriting standards is a concern as banks strive to achieve loan growth and to maintain or grow market share. Easing of underwriting standards in commercial, CRE, and auto lending presents increasing credit risk.” The OCC also said “rapid CRE loan growth over the past year and recent underwriting reviews raise concern over the quality of CRE risk management, particularly managing concentrations.” If the CRE sector has problems, elevated credit risk will mean elevated losses.

Now, a truly data-driven Fed, hiking due to incoming data and adjusting the medium-term outlook for employment and inflation, is one that hikes in a more unpredictable way anyway. Of course, if the economy is good, then this should lead to a steepening yield curve as term premia increase due to policy uncertainty, making long-lived assets relatively less attractive. The fact that the yield curve has been flattening though tells us that something is not quite right with this picture. To me, it speaks to a need for caution as the Fed normalizes.

Bottom Line: Nothing in the data says recession. In fact, recently Gavyn Davies wrote a good piece recently on why we should think of the global economy as accelerating. But there are signs like the curve flattening, weak Q1 GDPNow numbers, slowing consumer spending and auto credit delinquencies that suggest Fed hikes could crystallize credit risks. And given recent hawkish Fed statements and the potential for even more than three hikes, all of this is bullish for longer-duration Treasuries but much less so for auto ABS, high yield and emerging markets. Watch the oil price to see whether the market’s supply overhang reduces price enough to damage that credit market. If it does, we will see a major risk-off move in 2017.