For now, Americans are still working under the assumption that there will in fact be an election in 2020.
But assuming America is still a place where voters will be allowed to choose their leader, it’s natural to ask whether a prospective economic downturn tentatively “scheduled” for 2020 may end up doing what multiple investigations have failed to do: Undercut Donald Trump’s chances at winning a second term.
Trump of course campaigned on a promise that his policies would usher in a veritable economic renaissance. In pursuit of that, he’s piled expansionary fiscal policy atop a late-cycle dynamic – and all funded by debt.
That effort has met with varying degrees of “success”, but when it comes to GDP growth, Trump has come up a bit short, something he blames on the Fed. Here, we’ll let him explain:
In reality, it could be that a new economic paradigm wherein the traditional business cycle has seemingly been replaced by stagnation, means the “MAGA” promise (vis-à-vis the economy) simply isn’t feasible.
“My thinking is that this is probably the end of [the] traditional business cycle – no more big and ‘frequent’ amplitudes, but more like undulations around a flat line”, one strategist we spoke to this week suggested, adding that conversations about curve inversion and recession are thereby largely meaningless. There are no more recessions in the traditional sense. Just more or less of stagnation.
In any event, let’s say, for argument’s sake, that recent signals from the yield curve are reliable and that the traditional lag time holds true. In that case, the US economy may well enter a recession during an election year.
“IF the traditional 12m to 18m lead time between the curve leading growth materializes in this cycle, the economy will very likely be slowing materially come the 2020 US election”, Deutsche Bank’s Alan Ruskin writes, in a new note, adding that a “superficial view would be that the flattening curve will make it more difficult for the incumbent President Trump, or a candidate from the Republican party, to hold on to power in 2020.”
Here’s a visual that shows the 2s10s with a 400-day lead-time:
Ruskin then sets about answering the following question:
At a similar period in the past, what did the yield curve look like, and what did it say about the election result?
Spoiler alert: It is basically impossible to answer that question because irrespective of the inherent wisdom contained in the famous “It’s the economy, stupid” quotable, voters’ decision calculus is so complex that it simply isn’t amenable to attempts to isolate for one factor or if it is, that’s the purview of something that would appear in a peer-reviewed academic journal article, not a short piece of sellside research.
That obligatory caveat out of the way, this is most assuredly a worthwhile note from Ruskin if for no other reason than how topical it is. He uses four charts to illustrate his findings, each of which looks at a different time slice. Figure 1 is the most relevant. Here’s Ruskin to explain what it shows:
Figure 1 is the [3M-10Y] curve 18m before [an] election. The x axis shows the 10y minus 3m yield curve in basis points. The dots on the positive side of the y axis are labeled with the President’s name, and occurred when the President of the same party as the previous term was elected – i.e. the incumbent’s party won again. The dot on the negative side of the y-axis, are elections where a President from a different party to the incumbent President won the election. The red dot is the median for each occurrence – when the incumbent’s party won again, and when they lost.
Ok, so the straightforward takeaway is that the red dots (the medians) do in fact suggest that a flatter curve 18 months prior to an election “implies some greater propensity for the curve to foretell a change away from the incumbent President’s party”, to quote Deutsche Bank.
One standout is that there are only two instances of the 3M-10Y being inverted with 18 months to go until the election. As Ruskin notes, “on both occurrences… the economy was in unusually poor shape, almost certainly encouraging the Reagan 1980 and Obama 2008 wins.”
Invariably, some readers will criticize this as being impossibly muddy by virtue of all the other factors at play, but in a sense that’s actually what makes it interesting – or at least for those of you who enjoy this kind of thing. For instance, Ruskin goes on to document what contributed to instances where the incumbent party lost an election despite the yield curve being steeper than 200bps 18 months prior to the vote. To wit:
There are plenty of exceptions to any rule that a steeper curve 18m before an election, favors the incumbent’s party pursuing a successful Presidential campaign. As examples, all of Presidents Trump 2016, Clinton 1992 and Carter 1976 won against the incumbents party’s candidate, when the 10y minus 3m curve was steeper than 200bps 18 months before the election. Each of them had a special story that is worth noting: President Carter won partly as a backlash to an impeached Nixon; Clinton won, partly because of broken economic promises (‘read my lips’ on taxes, and, data that was ultimately better than appeared at the time); and, President Trump likely won partly because as much as the yield curve signaled growth was on a solid footing, income distribution issues had created considerable economic dissatisfaction.
What does this entail for Trump in 2020? Well, here again we come right back to a point we’ve been pounding the table on in multiple posts of late – namely that with fiscal policy exhausted and with the Fed having limited breathing room (don’t let it be lost on you that for all the “tightening” talk, rates are still very low and the balance sheet is still engorged), we may well be on the verge of seeing some “creative” policy “solutions”.
“Currently, the curve’s message is rather simple and suggests that come the 2020 election, the electorate may well be demanding new economic ideas in the face of a slowing economy, that in turn may produce more ‘out of the box’ thinking than in traditional campaigns”, Ruskin goes on to say, before exclaiming that “coming as it does against a backdrop of unusually large cyclically adjusted budget deficit for at least the next 5 years, this is otherwise, a far from propitious moment to ‘prime the pump’!”
Right. Which means Trump will invariably lean even harder on the Fed. And if that doesn’t work, he’ll likely resort to even more fiscal stimulus – deficits and discipline be damned.
Coming full circle to our “gentle” suggestion that 2020 may well be “different” in a way that, for now, Americans are understandably reluctant to ponder, we’ll simply close by asking whether it’s at least possible that if the economy does start to roll over next year, analysts and economists might start to notice some “discrepancies” creeping in between their real-time activity indicators and the “official” data….