FISCHER-ING FOR DECOUPLING
EDU DDA Jun. 2, 2025
Summary: Sweden is the latest to show negative GDP in Q1 even before getting to tariffs and the payback, joining the globally synchronized ranks of Japan, Britain, America. The bigger threat moving forward isn’t trade wars, it’s the distortion tariffs had already created from late last year. Comparing its effect this time around in 2025 to the supply shock in 2021-22 produces some eye-opening interpretations. In addition, the ISM’s manufacturing PMI for may suggests the payback hasn’t really started no matter how big while construction spending adds more cyclical clarity on weakness - if the data is good.
IT WAS A LOT OF INVENTORY, YET THERE’S MORE TODAY.
As Scandinavia echoes the re-recessions spreading around Asia in places like Japan and South Korea, “soft” economic data and cyclical views keep coming in on the negative side despite trade delays (that we need to recognize are already in danger of unraveling). In fact, Sweden’s trend is eerily similar to Japan’s; eerie because those are the two countries I always use to demonstrate the functions of a reserve currency if only to cement how globally synchronized is the more important factor than anything local.
For the US economy, many remain convinced of its invulnerability, however the ISM missed badly plus construction spending declined for the second straight month. The latter is potentially significant for its relation to labor market tendencies; if we can ever get some straight and dependable data on the segment out of the Census Bureau.
The greater and more immediate danger, in my view, is the payback. American imports soared – to put it modestly – in February and March only to then crash by a record amount in April. And it still isn’t anywhere close to enough to pay all the artificiality back. I don’t get the sense people have any idea how huge the distortion was, and therefore what that will mean. To correct that, we’ll look at and compare some numbers.
Overseas central bankers do seem to be “getting it”, which is why so many of them are abandoning their Economics (capital “E”) on “inflation”, opting one after another for more rate cuts while chasing the growing downside acceleration. Sweden is simply the latest of these.
This payback isn’t going to be exclusively an overseas risk, even if current commentary refuses to let go of “decoupling.” In times like these, those arguing for it might best remember what former Fed Vice Chairman Stan Fischer had admitted a decade ago. Fischer passed away over the weekend, and while he was an Economist through and through, he was also one of the few who seemed genuinely, if still entirely too infrequently, attached to basic economics – though, in the end, Fischer never was able to figure them out thanks, as always, to QE.
Truth in Scandinavia
There wasn’t a mainstream Economist, it seemed, who didn’t pass under Stanley Fischer’s tutelage. Fischer could count Ben Bernanke and Mario Draghi among his students, along with Larry Summers and Greg Mankiw. Even current BoJ chief Kazuo Ueda was on Fischer’s roster at some point at MIT (the utter damage that school has done to the world).
Why more of them didn’t share the small degree of openness the former Bank of Isreal Chief and Fed Governor did remains a mystery. This isn’t to say Fischer wasn’t conventional, he was. Only there were a handful of times when he broke with convention to question what the hell the Fed was really doing, if it actually was being successful and sharing that in common more with Janet Yellen than anyone before or since.
The funny thing was a lot of that introspection went on during 2014. Before giving Fischer too much credit, the timing wasn’t an accident. This was the year of the “best jobs market in decades”, what all the QE-ers had been waiting for since December 2008. In other words, policymakers thought they were safe to start reviewing the half decade of failure leading up to then, covered by a recovery they were thinking had finally shown up however belated.
It's far easier to conduct an accounting when you think you’re at the finish line. Yellen went on to admit in 2014 that the obvious “slack” in the labor market wasn’t just lazy, drug-addicted American workers after all, how failure of QE to stimulate a more robust labor rebound shouldered at least some of the blame.
In that same year, Fischer had jetted off to Sweden and then confessed:
Year after year we have had to explain from mid-year on why the global growth rate has been lower than predicted as little as two quarters back. Indeed, research done by my colleagues at the Federal Reserve comparing previous cases of severe recessions suggests that, even conditional on the depth and duration of the Great Recession and its association with a banking and financial crisis, the recoveries in the advanced economies have been well below average.
That was another favorite 2010s distraction employed among policymakers, deflecting from the clear weakness in “recovery” by stating how post-crisis recoveries tend to be weaker than most from the start. While technically true, as Fischer stated that still couldn’t account for the utter devastation in the world, and to workers most of all.
Notice, too, how the former Fed Vice also correctly put the entire advanced world into the same category. Yep, a little globally synchronized from a Fed official who thought it safe to begin stating some useful facts for once.
It didn’t last long, of course, especially as Euro$3 was already wrecking things in 2014 as Fischer and Yellen were thinking they’d achieved solid recovery ground. Instead, 2015 and 2016 pulled everything back, “unexpectedly” erasing the best jobs market in decades that, in all honestly, wasn’t really all that good. He never did figure out how it all got to be so synchronized, nor would Yellen. At least Fischer recognized some of the symptoms.
The main points, however, stand and very much apply to circumstances in 2025. Unanswered promises of soft landings and recoveries. Most of all why: globally synchronized. There is no decoupling, only a matter of timing and degree. If Fischer felt it necessary to apologize for the years leading up to the “best jobs market in decades”, what would he have done for the now-several years of “forgot how to grow” as it followed the pandemic?
Whatever Stan’s possible take, here we have Sweden revising its Q1 GDP to a negative (and its past GDP estimates a lot lower), putting it on course with Japan, South Korea, UK, Mexico, and the growing list of nations already showing contracting output even before reaching the tariff shock and full payback.
Not seeing clearly
Being unable to distinguish signals isn’t necessarily surprising; after all, the artificial high last year was widely considered a legit upturn, with many calling it a full recovery in the global manufacturing sector which prior to it had been sunk in a prolonged recession of its own. Just look back on all the commentary when the ISM Manufacturing reached fifty for only the second time in twenty-six months to start out 2025; the champagne was flowing all over the financial media, encouraged by Mr. Powell’s strong and resilient readings on the economy.
Everyone was fooling themselves over it; the phoniness behind it all was clear as day. It was no recovery, straight- up fake front running. Demand was being pulled forward. The only question then, and now, was just how much.
Standing at one extreme end of the scale, US imports have argued the artificial high was exceptionally high, therefore the payback following from it would necessarily have to be severe. And that’s before getting into any questions surrounding any more downside to consumer and business spending; in other words, just holding those constant, there’d still have to be an absolutely obnoxious downside to that high.
Because it was so enormous, without making some close comparisons to another period when imports (and inventories, the other part in all this) had soared to a similar enough degree, it’s difficult to really comprehend the scale.
While the earlier eruption was three and four years ago now, most of us clearly remember the supply shock – both parts. In the first half, goods were in such short supply inventories got run down to levels our modern sensibilities had never thought possible (the empty supermarket shelves, for instance). That meant as the snarls eventually loosened around the supply chain, there was a flood of imports as wholesalers and retailers began to catch up.
That second stage went even further owing to the preceding widespread shortages – remember the over-ordering? From late in 2021 throughout 2022, everyone in the goods economy was almost proud at how much they were ordering way beyond immediate needs. They weren’t going to let their customers down again, so demanding double, triple, quadruple the number of items became commonplace (for a short time).
By the numbers: starting with March 2021 (the final “helicopter”), American imports of goods from the rest of the world soared an incredible 32% (dollar values) by the end of March 2022, a period lasting thirteen months until the forgot-how-to-grow world began to emerge from that last oil shock. It wasn’t entirely the price illusion, either: real imports (adjusted for price changes) zoomed upward by almost 18%.
It was so big and overdone that the reversal from the middle of 2022 onward contributed much to the forgot-how-to-grow globally synchronized economy which took over from there. The goods recession kept on recessing for roughly eighteen months afterward, so that by the middle of 2023 real imports of goods into the US had slid 10% from that high. By value, the downside was almost 13%, leading to a period of agonizing retrenchment for way too many places around the globally synchronized world.
How about now?
In just the four months ending with March 2025, US imports surged by 23% when measured by both dollar value and price-adjusted volume. Yes, the artificial import high from the end of last year to a few months ago was even larger than the thirteen months of the entire supply shock. Both phases, the restocking empty shelves and also the overordering which followed.
If the downside from that earlier trend helped create and maintain the original forgot-how-to-grow world, what is the potential here in 2025 as even higher artificiality drains away?
It should be pointed out that some of the jump in imports, roughly $20 billion (value), was non-monetary gold, thus a distortion of a distortion. Even so, it was less than a third of the total import jump, meaning whatever there might have been in bullion the import craziness this time around was gigantic anyway, equal in terms to 2021-22, therefore being a sizable and increasingly immediate threat to the global trade in terms of demand payback.
Census’ initial estimates for April already crashed by a record amount, falling by 19.8% (by value) just from March’s total (the volume estimates aren’t released with the advance figures, so we’ll have to wait a few more days before they are published). Just going by the payback from the supply shock, there’s a bigtime downside yet still to come before even getting to timing and/or second derivatives plus second or third order effects afterward.
Maybe the most concerning aspect is that the real hit to production – anywhere – has yet to really develop. The forward-looking ISM, for example, only slipped a few tenths more (though it was supposed to have rebounded in May). And while it has erased that very brief “recovery” at the start of the year, the downside is just getting started.
ISM’s New Export Orders Index crashed to 40.1, while the imports version plunged to 39.9. Put these pieces together, what it says is that the contraction in the goods economy is still months ahead of us and that when it arrives this isn’t likely to be some small speedbump for a resilient economy.
If anything, this begins to account for way key financial measures like swap spreads haven’t budged from their April lows; most maturities remaining with a few bps of their records set during the deflationary liquidations.
Unsteady Census
Census reported another potentially important signal from construction in the US economy to go with everything else. I write “potentially” not based on the signal itself, that one has been repeatedly validated as a cyclical indicator in each modern recession cycle. The problem is the reliability of the data.
While residential construction does have a cyclical element to it, it’s actually non-residential construction which fits best with these inflections. And it makes intuitive sense, too. Companies facing a serious, more-than-temporary downturn will take defensive steps to protect their balance sheets.
Capital expenditures are one discretionary line item that business can exercise control over. It often requires substantial planning and financing, therefore matches the longer run parameters of business cycle changes. Quite simply, when business management grows pessimistic beyond some threshold, they naturally pull back on capex preferring to better time investments with perceived opportunities while most importantly savings corporate resources.
It’s the correlation with the jobs market that really gives the data its power and, again, for intuitive reasons. Companies that are serious about negative prospects aren’t just pulling back capital plans, they’re very likely limiting if not eliminating hiring at the same time considering if not pulling the trigger on layoffs. There is a more than decent correlation between non-residential construction spending and data like the Establishment Survey’s payrolls (and hours).
What the Census data shows is a more-than-short-run rolling over process for the economy once it enters the cyclical inflection. Construction spending rises quickly during expansions (or just rises, as in post-2008 reflation), begins to tail off - meaning sideways - before contracting during concurrent to the general downturn phase.
While the process is clear, the data not so much. The series is subject to often wild revisions in the short run; I’m not even talking about benchmark changes which have taken the rewriting several steps beyond. In short, it’s tough to get a confirmed signal from what otherwise is a reliable cyclical indicator.
As it currently stands – being aware of the revision potential – the construction series has declined for two months running (typically a cyclical warning) which makes three out of the past four months. And that follows an extended period of sideways, raising the chances of a cyclical inflection. If these estimates are not revised away over the months ahead, that’s possibly confirmation for an economy going beyond forgot-how-to-grow.
The mainstream will surely look at the construction data and the potential contained in it as a product of trade war “uncertainty”, that companies are holding back now thanks to tariffs and a business climate they no longer have any clarity over.
The economy, however, keeps getting more not less certain. The negative signals have been piling up since last summer, and that includes the artificial high – in fact, the most alarming result in everything is how high that artificiality may have attained. That downside remains ahead of everyone and, if anything, it has yet to really begin impacting even the farthest reaches of the supply chain which are still benefiting, so to speak, from the positive sides to the distortions.
And while that may seem like a “them” problem for everyone else around the world, as Stan Fischer once very briefly admitted, there is no real decoupling.
