The Uncertainty Machine

POSIWID

Stafford Beer, source: By Universitätsarchiv St.Gallen | CC-BY-SA 4.0, CC BY-SA 4.0, https://commons.wikimedia.org/w/index.php?curid=141494297

In his very good book, The Unaccountability Machine, author Dan Davies spends a fair bit of time discussing Stafford Beer’s heuristic, “The purpose of a system is what it does”, or POSIWID.

Adopting a POSIWID analytical approach means that instead of focusing on the expressed intentions or desires of any large, complex system or organisation, you look at what it actually does and accept that what it does is indeed its purpose.

Given company executives and governments worldwide are looking for answers and certainty about what Trump will or will not do on trade and tariffs, I’m starting to think POSIWID is probably useful in this context, too.

Or, to put it another way, if everything Trump does and has ever done creates uncertainty, then we should all assume that creating uncertainty is indeed his ultimate objective, whether he knows it or not.

To give an illustrative example from Trump 1.0: Steel tariffs and Canada.

Last time round (thank you, PIIE, for the timeline), Trump started an investigation into the national security threat posed by steel and aluminium (April 2017), announced tariffs on imports of steel and aluminium from Canada and others (1 March 2018), announced a temporary reprieve for NAFTA countries (8 March 2018), extended the reprieve for Canada and some others (30 April 2018), ended the reprieve for Canada, Mexico and the EU (1 June 2018), removed the tariffs on Canada and Mexico (17 May 2019), reimposed some tariffs on Canadian aluminium (6 August 2020), and finally ended the tariffs on Canadian aluminium but demanded quotas instead (15 September 2020).

Phew.

Even if you think you’ve got a deal with Trump, maybe tomorrow he changes his mind. Maybe tomorrow, he wants something else. Maybe tomorrow he forgets. Maybe tomorrow he decides he doesn’t like you anymore. Certainty doesn’t exist.

At best, what you have is a temporary reprieve. But what you will never get is an end to uncertainty.

So, if you’re a company facing down The Uncertainty Machine, what do you do?

You could …

1. Get on with your life

Given you can’t know what will happen, you could just get on with doing what you were planning to do anyway. Sure, you go through the motions — bend the knee when needed, donate to the odd thing, make sure not to say anything negative publicly — but otherwise, you continue to trade and invest as you would have done otherwise. If tariffs happen, then you deal with it then.

Que Sera, Sera.

This is absolutely the approach taken by some. But it is probably a function of relatively low exposure. Either because the things they sell aren’t particularly price-sensitive, exposed to tariffs (e.g. a service), big in the US, or originate in countries that Trump doesn’t talk about very much.

2. Freak out

Uncertainty is very bad and scary, and at any moment he could do something that wipes out your market share and PANIC.

In practice, companies reacting like this are usually in the opposite position to those in #1, and it usually doesn’t last for too long before they try to do something, anything, about it. Even if something or anything doesn’t achieve very much.

3. Remove yourself from the game.

Given tariffs could land at any point and the uncertainty is killing you, you could make sure you’re not exposed to US tariffs by either a) only selling stuff to Americans if you’ve made it in the US or b) stop selling things to Americans.

Both have their complications:

a) Only sell stuff to Americans if you’ve made it in the US

Until recently, I would have said that this is a really good option, assuming you’ve got the money to invest and the consumer demand justifies it. You can’t be tariffed if your goods never have to cross the border into the US **taps forehead knowingly**. You could even argue that this outcome is what Trump is actually trying to achieve through his imposition of perpetual uncertainty.

And you know what, for most companies, this probably still works. But not all.

The recent US decision to block Japan’s Nippon Steel from acquiring US Steel, despite, from the outside, the deal being in the interest of all involved, calls into question the “invest in the US” option. While his position could change (see: everything above about uncertainty), Trump also opposed the takeover.

There are also questions in other industries. For example, there is a plausible scenario in which Chinese EV manufacturers dodge the massive (currently 102.5 per cent, could go higher) tariff by investing in the US and making the cars locally. The same could be true of the supply chain, particularly the batteries. On paper: more American jobs, more investment — everyone is happy.

Buuuuuuuuut, new regulations mean it’s not quite that easy. The new US-connected vehicles rules, which I wrote about earlier in the year, went live last week and, once fully in force, will essentially prevent any Chinese EVs from being placed on the US market, even if they were made in the US.

Could Trump decide to exempt Chinese EV companies that invest in the US from these rules? Probably? Will he? Who knows.

b) Stop selling things to Americans

I mean, if you don’t have to, don’t. I guess. For lots of companies, this isn’t an option, but for some, it is.

4. Play the game

Look, there’s always the chance he’s going to turn on you. And a deal is only a deal for so long as he decides there’s a deal. But a temporary reprieve is better than nothing. And who knows, you might get lucky! So go all in, full MAGA. You’ve always loved him, he’s always loved you … until maybe he doesn’t anymore.

5. Insurance

This is more of an afterthought, but surely, if you’re doing any of the above, you could buy insurance. Every company does this when faced with this kind of uncertainty.

Now, I don’t work in financial services, so maybe Trump Insurance already exists, or maybe it doesn’t because it’s a really stupid idea, but surely someone somewhere has created a product where company A pays x amount a month to company B as a regular premium. If there are no new Trump tariffs on the product company A exports to the US, then company B has a nice steady stream of income. However, if Trump raises tariffs above an agreed threshold, say, 20 per cent, then company B has to pay company A an agreed amount of money.

Easy! The best thing about a product like this is that you could then put an actual price on the Trump-related tariff uncertainty and even trade it. Companies could sell the insurance and then lobby Trump against imposing tariffs on the things they insure. We could create a new tariff-industrial complex. This would be great fun for everyone involved and definitely wouldn’t go horribly wrong when The Uncertainy Machine wakes up one morning and decides to put a 200% tariff on all imports into the US.

---

Chart of the week

Sander Tordoir and Brad Setser have published a new paper looking at the impact of the second China shock on the German economy.