One Bank’s Clients Complain ‘The Sheer Volume’ Of Tweets Makes It Impossible To Hold Positions

One Bank’s Clients Complain ‘The Sheer Volume’ Of Tweets Makes It Impossible To Hold Positions

The weekend brought further evidence to support the contention that the Chinese economy remains fragile in the face of mounting uncertainty.

The official manufacturing PMI printed 49.5 for August, a touch below consensus, and consistent with the idea that any upticks will likely prove fleeting. New export orders moved higher but remained in contraction. The internals were generally weak and appear to support the case for more policy accommodation, especially now that the trade war has escalated anew.

Obviously, the more contentious things get between Donald Trump and Beijing, the greater the scope for sentiment to weaken further as businesses dig in and brace for the worst.

Predicting the next plot twist is synonymous with predicting what Trump will tweet. In other words, forecasting the next turn is impossible. That’s a nightmare for businesses and investors alike. Throw in the US president’s monetary policy tweets and it’s a total crapshoot.

“A common discussion point with clients recently has centered around the difficulty, or even inability, of holding onto positions for extended periods of time given the sheer volume of – sometimes conflicting – messages by US President Trump on issues like US-China trade policy and the Fed, arguably the two most important themes driving markets at the moment”, Barclays wrote Sunday, in an amusing little bit called “A little (blue) bird told me…”

Currencies react more to Trump’s trade tweets than Fed pronouncements, the bank went on to say. Specifically, Barclays divided the president’s 2019 tweets into those containing “Tariffs and China” and those containing “Fed”. That exercise produced 17 distinct trade tweets and 26 tweets criticizing Fed policy. They then looked at EURUSD, dollar-yen and the yuan from 10 minutes prior to the tweets to 20 minutes after. Here’s the return distribution:


Clearly, Trump’s trade tweets have a larger impact than his tweets about US monetary policy and it’s not hard to discern why.

“The lack of reaction to Fed-related tweets may reflect that they are entirely one direction – criticism – that is well-known and lacks new information about policy direction and relatedly that the effect of such criticism more likely has long-run corrosive effects undermining market faith in US institutions that are hard to observe in high-frequency USD changes”, Barclays writes of their findings.

That latter bit is important. Although Trump’s Fed tweets are now so common that the market tends to ignore them (or most of them anyway) in the near-term, over the longer-run they may serve to erode the market’s confidence in the US government and the dollar as an anchor in a world of uncertainty. That’s entirely consistent with the fact that the US government is now the source of that uncertainty.

The bank goes on to note the obvious, which is that “trade-related tweets have been both escalatory (risk negative) and conciliatory (risk positive) and contain more immediate information about the direction of policy”.

Of course, even if you can draw an ostensible distinction between Trump’s Fed tweets and his trade tweets, they are really just two sides of the same coin, something Barclays readily acknowledges.

“As the Fed has clearly signaled, trade policy is having an effect on their policy path as well, meaning that trade tweets indirectly may also contain information about Fed policy”, the bank says.

Little wonder, given how tangled and circular all of this is, that the bank’s clients are having trouble (an in some cases finding it wholly impossible), to hold positions for any length of time. It’s just a reflection of Trump’s own inability to take a position on the issues and stick with it.


2 thoughts on “One Bank’s Clients Complain ‘The Sheer Volume’ Of Tweets Makes It Impossible To Hold Positions

Speak your mind

This site uses Akismet to reduce spam. Learn how your comment data is processed.