This week we again asked an AI to generate a market summary and, having checked that it was getting its data from various reliable asset managers and media outlets, it did a great job of summarising about 30 pages of commentary in around 700 words almost instantaneously. We’d even go as far as saying that it did a better job of simplifying some of the issues than most finance writers (present company included). You can download the text here if you are interested but for the purposes of this column we will, perhaps counterintuitively, paraphrase the AI in the first instance – US inflation moderated, the Federal Reserve temporally paused its rate hiking cycle while consumer sales and sentiment gauges firmed. On the face of it, this looks like an immaculate ‘disinflation’, and the dominant narrative in the press is that a resilient US consumer has fanned hopes of a soft landing.

The new news in markets last week was upcoming stimulus by the Chinese authorities, and that was probably the thing that moved markets the most last week, as bad economic news (a weaker than expected Chinese economy) became good news for markets. Markets probably had known that inventories have been rising in Asia, and manufacturing in China and in the West is clearly slowing, but we didn’t know what the reaction of the Chinese authorities would be. There could be some truth in both, but the reaction of commodity markets and many China bellwether stocks (like French luxury goods maker, LVMH, and host of other European exporters) suggests the latter was perhaps more influential. This was also reflected in that other Chinese bellwether, the Australian Dollar, while the Australian stock market also seemed to breathe a little sigh of relief, with all sectors and most stocks up a few percent. Otherwise, the Nasdaq continued to outpace, driven by anything remotely AI related, and Japan also continued to be a stand-out performer (possibly also buoyed by news that its largest trading partner was about to stimulate its economy).

So far this is a matter of emphasis, and the salient fact is that markets are up across the board. However, the more you look at the current data the noisier it gets, the more one feels a bit sorry for central bankers, and the less useful an AI generated market summary feels. Take the two ‘issues du jour’ – inflation and the potential incoming recession. The two charts below are on inflation. The one on the right shows monthly US CPI vs Core CPI (ex-Food and energy). Both oscillated around a 0.125% a month (or 1.5% a year) pre-COVID, and then they both got very volatile. Now falling energy prices and easing supply side restrictions mean the overall number is going down. Meanwhile, the arguably more important core number seems to have settled around 0.4% (5% a year). Current inflation data gathering techniques are notoriously archaic and lag quite a bit, while the chart on the left shows another more recently published (and one might say experimental) measure called Truflation, which uses real time sales data. It’s early days, but it (and its underlying constituents) implies that inflation is actually plummeting and is almost down to the target of 2% already.  This latter measure is corroborated by real time rent data, which is a large part of CPI, but then again, the so-called super core (ex-housing, food, and energy) as well as other anecdotal data suggests services inflation is indeed persistent in the US. Meanwhile, the UK’s inflation problem is getting worse, and data poor Australia wonders whether it is going to have a UK or US type experience. In that light, it is little wonder that Jerome Powell and the Federal Reserve sought to buy some time by pausing the rate hike cycle but talking up the prospect of future rate rises last week

Turning to the probability of a recession, manufacturing data has been as unequivocally weak as the US Consumer has been robust, and the following chart sums up the mixed messages coming from bond markets. The New York Fed model that predicts the probability of recession relies largely on government bond prices, and where markets think rates will go, to predict the probability of a recession. At anything above 50% it has proved pretty prescient, and it has never been this sure of itself (80%) since the purposefully self-inflicted ‘recession we had to have’ in 1982. Meanwhile, high yield corporate bond spreads (in green) measure how much extra yield compensation investors require in case the riskiest corporate borrowers default. These are at just above average levels and are apparently heading down, while implied default rates are at historical lows.

We always knew it was going to be a noisy period, so we think it is probably best to take all this data with a pinch of salt, stay alert rather than alarmed, and be ready for something to emerge from the fog.  Next week there will be some important US housing data, and Jerome Powell might reveal a bit more about the Fed’s thinking during his semi-annual grilling by the US Congress. Also, we should get a better feel for how far China will go in stimulating the world’s second largest economy. However, realistically, it will probably take another couple of months and another results season before much of this is resolved.

‍