Sure, you might think: Donald Trump isn’t exactly a competent president. But it’s a long-standing truism of U.S. politics that, at the end of the day, presidents really don’t have immediate and severe effects, for better or worse, on economic performance or jobs. Instead, what really matters are larger-scale forces — say, the growth or stall of productivity, something that politicians have very little effect on in the short term. We can all play games with economic statistics and where presidencies begin and end, but most of the claims involved are partisan fictions. 1

But that truism was never tested by Donald Trump.

Few seem to have adequately priced in the possibility of large, unusual downside risks from having Trump in the White House. I’m not talking about normal policy differences, such as Trump’s withdrawal from the Paris climate deal, in which some will argue (just in terms of economic development) that he’s freeing U.S. businesses while others will maintain that focusing on coal mining while the future is in renewables is a poor trade-off. I’m focused here on the possibility that his chaotic presidency could produce devastating results just because normal governing might prove impossible.

Here are the five biggest scenarios I’m aware of, and how the chances of each have changed since Trump won the presidency in November. 2

Trade War (unchanged). Perhaps the most obvious one, and the one that is most policy-based. Presidents have quite a bit of leeway on trade, and if Trump acts on his clear policy preferences he still could find himself (and, more importantly, the nation) in a spiraling situation with no good answers and with immediate economic consequences. On the good side, Trump hasn’t actually done anything destructive in this area, and he’s made it clear to everyone, foreign nations included, that most of his threats are just bluster. But on the bad side he’s continued to use trade-war rhetoric — most recently against Germany — making it even more clear now than when he was elected that he really holds this policy position, and therefore may eventually act on it.

Government default/shutdown (increased chance). The debt limit will have to be raised sometime this summer or fall or else the government will default. Funding bills are needed by the end of September to keep the government running (in both cases, temporary extensions are possible). It is extremely difficult to get to an extended shut down of the government or, even more dangerous, a default on the government debt payments. The former has only happened twice; the latter has never happened, and neither has ever come particularly close during periods of unified government. So the chances are still low, but I’d have to say that dysfunction in the House of Representatives and the White House so far this year has made the chances of one or the other a real, if not yet too large, possibility. And getting close could damage the economy even if the threat is averted at the last minute.

The president causes economic damage by saying something crazy (decreased). Here’s one where Trump’s widespread reputation as a paper tiger helps. His ability to spook the stock market or otherwise harm the economy with some stray remark is probably almost entirely gone; already by February observers were noticing that he could no longer move individual shares with his tweets. It’s perhaps still possible that Trump could spark a foreign policy crisis by saying something inappropriate, but even there most foreign nations have probably learned that much of what he says can be ignored.

Executive branch mishandling a crisis (increased). How would this administration deal with a Katrina-like natural disaster, an epidemic, or a financial crisis similar to the one in 2008? Well, they would apparently throw a bunch of empty desks at it. Take the Department of Treasury. The nominee for Deputy Secretary has withdrawn, eight nominees are waiting for confirmation, and several other important spots such as undersecretary for domestic finance still have no nominee. In an emergency, not only would the missing people be a risk factor for the smooth functioning of the department, but coordination among multiple agencies would be especially difficult. An inexperienced and error-prone White House wouldn’t help. We knew back in November that Trump himself had no government experience, but the rest of this looks a lot worse now than it looked then.

The president mishandling a crisis (unknown, but probably increased). Even a well-staffed executive branch, with a highly functional White House, depends on the president himself or herself to lead when some external shock requires coordinated government action. Only the president can fully command the attention of executive branch departments and agencies, pushing them (if needed) to drop other priorities and focus on a crisis. Only the president has the bully pulpit to focus the attention of the nation, if necessary, on something that must be done. So it can really matter whether the president has good judgment, knows how to use technical and political advisers, has the respect of others inside the government and out, and knows when to act and when to allow others to act. Perhaps, if and when some crisis shows up that Trump didn’t cause himself, he’ll prove himself capable. So far, there’s no evidence of that, but then Trump remains untested by any serious challenge of this type. That’s unlikely to last.

Again: None of this takes into account the likely effects of deliberate policy, such as (for example) the likelihood that Trump’s jingoism is already harming the economy by depressing tourism and deterring foreign students from studying in U.S.

Overall? There’s very little good news here. That doesn’t necessarily mean disaster is around the corner. But it does probably mean that the downside risks to the economy from the current political situation are underappreciated.