Author: business

Stay informed with free updates

Simply sign up to the Central banks myFT Digest — delivered directly to your inbox.

The writer is a former professor of economics and senior adviser at the Bank of England.

There are many dark visions of the far future of artificial intelligence: Skynet vs humans, as depicted in the Terminator franchise; Nick Bostrom’s Superintelligence running amok, turning the world into paper clips; or a world of worklessness, with either poverty or abundance — ‘fully-automated luxury Communism’ (depending on who owns the robots). 

But a recent paper — The Emotions of Monetary Policy — which uses modern machine-learning techniques to crunch speeches at European Central Bank press conferences by Mario Draghi and Christine Lagarde, raises the prospect of another future that might not be all that far off, one that is also not wholly positive. I don’t know whether there would be a watchable sci-fi finance movie in it (please throw me a few $ as an executive consultant if you steal the idea), or even a new series of Industry, but read on and make your own judgement.

The ML method used in the paper, very roughly, is this: first, algorithms are run to turn video of facial expressions, the tone of voice in the audio, and the content in the text into data on emotions of different kinds. Second, these measures of emotions are compared with financial market data released during and after central bank speeches.

The inference of the paper is that the non-verbal information the algorithms are picking up (including smiles and frowns) do move financial markets. Whether this is causation or correlation, we can’t tell for sure. Maybe ECB watchers decode the non-verbal information informally. Or perhaps the correlation comes about because the non-verbal information eventually becomes verbal.

I don’t know how robust this work will turn out to be. The merits of this particular paper are not the main issue: there is a growing literature of work like this and looking beyond. As research like this becomes more widely known and replicable, and improves, one can imagine central bank watchers devoting resources to decoding visual and tonal information in real time. Perhaps algorithms used by ECB watchers could decode the non-verbal information better than the watchers were doing previously and informally. Computers can work flat out and don’t need to let off steam writing jokes in the Bloomberg chat.

The next step in the escalation of the algorithm war is that central bankers, knowing that their press conference speeches will be fed into ML algorithms, start to try to moderate their non-verbal communication to make sure that it has the effect on financial markets that they want it to. Perhaps they run rehearsals of their speeches through algorithms of their own, built to replicate the watchers’ programs as closely as possible, to try to gauge in advance how the watchers’ algorithms will judge what they say. Imagine a monetary policy maker talking over a small bank of screens reporting a guess of what the audience’s ML programs will be saying via 21st-century graphics, so that he or she can make adjustments in real time. Or just some equivalent of a producer in the speaker’s ear, saying “Ms. Lagarde, you need to smile a bit less” and “raise the eyebrows”.

The arms race then proceeds thus: The watchers go back and devise new algorithms that can rinse the new algorithmically informed speeches, themselves engineered to trick the old algorithms. Central bankers realise this and are forced to up their game. And so on.

Presumably the algorithms are going to produce mistakes: inferences about views about the economy and future policy that are not held by policymakers, causing (for example) yields to rise. These unintended effects then have to be responded to by more communication, those communications are gone over by the algorithms again, and the cycle continues.

Who knows where this ends? Perhaps central bankers withdrawing, either to reading pre-scripted remarks, or pulling back even further, reverting to releasing (AI-vetted) texts on their websites, lest reading the words out conveys something that was not intended and could unintentionally influence asset prices. In effect, that would mean computers taking over both sides of the press conferences, the speaking and the listening.

If central banking was a more definite science — if everyone on either side of the table understood what was producing inflation and exactly how to control it — there would be no need for thousands of processors to digest central bank speeches. But in that world, there would be almost no need for human policymakers at all. 

If central bankers were completely transparent and not deceitful, ambitious, fearful of being caught out, prone to disagreeing with each other, or forgetful, there would be little for algorithms to unearth. Everything we needed to know would all be out there in the policy documents, charts and forecasts. But then humans are not like that. It’s because of these faults that we form committees of them, to average out their frailties, and tie them in institutions that constrain them.

It’s not hard to see that these kinds of techniques could complicate all kinds of sensitive public speaking. Finance ministers delivering or being interviewed about a new fiscal plan, concerned about the impact they will have on the cost of financing that plan; CEOs of public companies giving a speech accompanying a new earnings forecast for shareholders, hoping to achieve as favourable a reception as possible. Political leaders negotiating with each other trying to lie and lie-detect. Parliamentarians listening to their leaders or vice versa. Corporate lawyers duking it out over the terms of a merger or acquisition.

Indulge me for a moment as I assume that humans are still doing the substantive bit of policymaking or business at some point far into the future, and have not been put out of a job by artificial intelligence programs. Perhaps the real AI singularity we need to worry about is not a solar system of paper clips, but an arms race of algorithms deployed by speakers and listeners in all public arenas, wasting vast quantities of computing power that could have some more productive use.

Unlock the Editor’s Digest for free

Roula Khalaf, Editor of the FT, selects her favourite stories in this weekly newsletter.

There was, briefly, a risk that FT Alphaville’s recent Art of the Chart show would occur without any small caged mammal representation.

But the arc of the mammalian universe bends towards content, and so — thanks largely to the Herculean efforts of MainFT graphics éminence Alan Smith — this graphic exists (zoomable image here):

Long-term fans of our scm coverage may recognise this is basically a fancier, mappier version of the extensive chart set we made last summer. The idea was to show the weird behaviour of prices recorded for a “small caged mammal”, one of the items the Office for National Statistics observes to track inflation, at individual shops, by showing every shop as its own line chart.

And, yeah, those lines are often weird — which suggests to us that despite some guidance (more on that here)…

…small caged mammal remains a slightly nebulous item type.

However, we have a confession. In the process of preparing the data for Sir Alan, we found our system of coding the price series of individual shops contained two material errors.

blunder 1

The first one went like this: in the ONS’s price quotes tables (aka the best data sets ever), each price comes with metadata including the region in which the vendor is based, and a unique shop code. The ONS’s guidance on how to use this data can be found here (nb that’s a direct download link).

When we made our shop-by-shop series, we gave each a name that combined its region (eg London) with its shop code (eg 023), leaving us with a set of shop like, uh, “London-023”.

We thought this sorted all our problems…

…but in our arrogance and conceit, we failed to realise that shops with the same code could also be distinguished by whether they were an independent (with 10 shops or fewer) or a multiple (with more than 10 shops). So, for instance, there could be London-023-Independent AND London-023-Multiple.

In retrospect, the fact that we’d made an error was obvious, because some of the charts showed prices that went beyond the actual highest price, a result of us summing two prices when an independent and multiple had the same region and code. It truly sucks to suck.

As far as we can tell, region, shop type and shop code are the three categories that determine a unique shop, so we think we’ve now completely fixed the issue — but how many shops did this effect?

By our count… three, which is to say we accidentally merged six shops into three: London-035, Scotland-016 and Scotland-029.

Unfortunately, though perhaps expectedly, all three are among those we then called out for having unusual price movements…

…and narrativised thusly:

— London-035: [2020–2024] Davide had been on the ONS small caged mammal beat for a while now, and still wasn’t really sure what he was looking for.

— Scotland-016: [2020–2024] Drawing a six-sided dice from her coat pocket, Anita prepared to determine what kind of mammal she’d look at this month.

— Scotland-029: [Late 2019] “We gotta switch things up. The sub-£45 mammal market — that’s where the money is.”

Fixing our error — and therefore splitting these up appropriately — the charts instead look like this:

It certainly makes London-035 look a bit more reasonable (although it still appears a marked chinchilla shift occurred at 035-M after the UK’s initial Covid-19 lockdowns), and Scotland-029 not longer has the severe price drop at the start. Scotland-016-M is still pretty weird.

blunder 2

Blunder 1 is stupid, but we’d dare to suggest understandable. Blunder 2 is a little harder to excuse.

Y’all, we erased Northern Ireland.

As mentioned, all these prices come with metadata including region. The regions are coded 1 to 13, where 1 is “catalogue collections” (we’re not sure how literally to interpret this), and 2 to 13 are regions of the UK. Northern Ireland is number 13 in this collection, Scotland is 12.

And, while converting the data to make our silly little name strings, it seems we converted both 12s and 13s into “Scotland”, dragging 10 Northern Irish shops into the Caledonian batch.

We would like to take this opportunity to apologise to anyone who has ever been associated in any way with Northern Ireland. Liam Neeson, Nadine Coyle, Kadhim Shubber, the Derry Girls… we’re sorry.

As far as we can tell, this will only have affected the actual line for one previously-presented shop, “Scotland-045” — which did exhibit a slightly odd uptick, and should in fact have been one NI shop and one Scottish one:

small caged mammal redemption?

Thankfully, these errors all got caught BEFORE we embarrassed ourselves by sending bad data to Alan and the graphics gang. The chart show map pictured above included the correct (we think!) series, and actually reflected all the members of the United Kingdom. Wahey.

But Northern Ireland deserves the representation we cruelly denied it last summer, so here’s all the little charts again, fixed and running up to the end of 2024:

We can’t wait to find out what fresh error we made while doing this. 😌 Data journalism, as ever, is hard.

Further reading:
— The UK’s inflationary basket case

Stay informed with free updates

Simply sign up to the US inflation myFT Digest — delivered directly to your inbox.

Egg prices in the US have finally started to drop.

The national wholesale price of eggs (sold loose by the truckload, in this case) dropped 15 per cent in the week ended March 7, according to the US Department of Agriculture’s latest report. It’s been a couple of weeks since the last big avian-flu outbreak, and about a month since the press and lawmakers started making more noise about prices.

Now, our readers can be forgiven for thinking that weekly moves in the price of eggs is not a huge deal. They would be wrong for thinking this, but we can understand.

Maybe they didn’t see John Burn-Murdoch’s work about how global inflation led to unprecedented global turnover in governing parties. Or maybe they’re not interested in food-supply systems or concentration among US agricultural firms. Or maybe they understandably like to eat animal products without thinking too much about where they came from.

But from a purely macro perspective, today’s Consumer Price Index data gives us this preposterous chart:

The average cost of a dozen eggs rose 12.5 per cent on a monthly basis in February. (It’s still 10.4 per cent seasonally adjusted.) That’s a 59 per cent annual increase. This, again, is preposterous, and has brought plenty of attention to the market.

Antitrust lawyer Basel Musharbash argues in a fascinating in-depth series for BIG that industry concentration (at all levels of the market) has removed competitive reasons to rebuild flocks quickly after outbreaks of avian flu.

Another commentator argues in the Atlantic that eggs should be expensive, in part because they’re delicate and break easily. (??) But a different argument in that piece — that cheap eggs require the “immiseration” of hens — conceals a very reasonable point. Factory farms are what churn out the lowest-cost eggs. And because hens are crowded into pretty close quarters in conventional cages, they’re more vulnerable to avian flu outbreaks.

This isn’t stated outright in the USDA’s latest market figures about the avian flu outbreak, but the figures do imply it:

The 30.3 million birds lost included 22.4 million (74%) in conventional caged systems, 7.9 million (26%) in cage-free systems, and 26,000 organic (0.1%). These losses represent 12.3% of the conventional caged layer flock, 7.8% of the non-organic cage-free flock, and 0.1% of the organic flock on January 1, 2025. Compared to January 1, the caged flock on February 1 was down 7.3% while the cage-free flock increased 1.2% and the organic flock 0.6%. As of the end of February, nearly three-fourths of layer losses have occurred in caged systems.

So it’s worth noting another announcement that preceded the decline in egg prices: The US government is spending $400mn “to indemnify producers whose flocks must be depopulated to control the further spread of [avian flu],” and to “aid farmers to accelerate the rate of repopulation, including ways to simplify the approval process to speed recovery.” Meanwhile, the US is cutting grants for cancer research.

Unlock the Editor’s Digest for free

Roula Khalaf, Editor of the FT, selects her favourite stories in this weekly newsletter.

After strong charts quizzes from Louis, Bryce and Alex, with lots of interesting squiggly lines and strong efforts from readers, the baton has passed back to me. Apologies in advance.

The rules remain the same though: identify the three charts below, email your answers to [email protected], putting “Quiz” in the subject line. We’ll draw one winner randomly from the pool of correct guesses that arrive by lunchtime Monday, and they’ll get a lovely I ❤️ Charts T-shirt.

We usually name everyone who gets all three correct so let us know if you want to be anonymous or pseudonymous.

Bonne chance.

Steven Kamin was previously head of international finance at the Federal Reserve and is now senior fellow at the American Enterprise Institute. Mark Sobel was previously head of international finance at the US Treasury and is now US chair, OMFIF.

In recent weeks, the buzz has been mounting about a new American plan — a “Mar a Lago Accord” — to upend the global monetary system. We can only hope it remains idle chatter.

In brief, based on a detailed discussion paper by CEA Chair nominee Stephen Miran, the accord would have America’s trading partners help weaken the dollar and commit to providing low-cost, long-term financing to the US government, enforced by the threat of higher tariffs or removal of security guarantees.  

Intriguingly, there has been no announcement by the Trump administration or even a tweet by Trump, but Miran’s paper — along with various utterances by Treasury Secretary Scott Bessent — have led Wall Street observers to believe such an initiative is indeed in the offing.

And that’s too bad, because a Mar-a-Lago Accord would be pointless, ineffectual, destabilising, and only lead to the erosion of the dollar’s pre-eminent role in the global financial system.

The Mar-a-Lago Accord is premised on the view that the dollar’s global dominance is bad for America. Unnatural demand has caused gross overvaluation. This has in turn led to reduced export competitiveness, persistent trade deficits, and the erosion of US manufacturing. In response, an Accord would call for the US and its trading partners to intervene in foreign exchange markets to sell dollars for foreign currency in a bid to get the dollar down.

However, since foreign sales of US Treasuries and prospects of dollar losses could push up US interest rates and jeopardise the financing of federal budget deficits, foreign governments would have to increase the duration of their remaining holdings of Treasuries, even buying 100-year zero-coupon bonds from the US government — in essence, free financing for a century! And because they could not be expected to do this voluntarily, they would be threatened with higher tariffs or the loss of American military support if they failed to comply.

So, what’s wrong with all that?

First, contrary to Miran’s view that the dollar’s global role is harmful for America, it is actually a net plus, facilitating our business activities abroad, lowering the cost of capital, and increasing our geopolitical reach. And even if the plan succeeded in lowering the dollar, it would do nothing to help the US economy or its workers.   

Much of our trade deficits reflect a buoyant economy and large fiscal deficits, not the strong dollar. Moreover, our trade deficits aren’t really a problem per se. Despite them, US economic growth has outstripped that of our major trading partners, and the unemployment rate is only 4 per cent — very low by historical standards.

In fact, there’s no logic to the notion that all countries should have balanced trade. We need trade deficits in order to provide an outlet for spending that otherwise would show up as economic overheating and inflation.

Moreover, the strong dollar clearly isn’t the cause of the shrinking share of US workers in manufacturing (now less than 10 per cent of total employment). The same trend has been at work the world over, in countries with both trade surpluses and deficits, on account of the rapid productivity growth in this sector.

Second, the plan would not succeed. As countless studies have shown, pushing the dollar down on a sustained basis would require the Federal Reserve to lower interest rates and foreign central banks to raise rates; but with US inflation stubbornly exceeding the Fed’s 2 per cent target and foreign economies languishing, that’s not going to happen.

By the same token, if foreign governments were busy selling Treasury bonds in order to depress the dollar, it’s unlikely that increasing the duration of their remaining dollar bonds could be enough to keep US interest rates from rising. And while threats of higher tariffs and ejection from the security umbrella might coerce Japan and Europe to play ball, China — which should be America’s main concern — is going to be less willing to kowtow to Trump.

Third, a Mar-a-Lago Accord risks undermining the global dominance of the dollar. That dominance is based not only on the safety and liquidity of US Treasuries, but also on the long-standing historic prudence of US economic policymaking and its support for a stable, rules-based global trading and financial system.

Mistreating our allies, breaking trade agreements, and undermining support for global institutions, as is now under way, will only encourage other countries to seek alternatives to the dollar. Trump has threatened countries with tariffs if they abandon the dollar, but nothing could accelerate that process more effectively than reckless actions against our trading partners.

Finally, an effort to force a Mar-a-Lago Accord on resistant trading partners could trigger a global financial crisis. The stock market is already in freefall on account of Trump’s capricious tariff policies. Consider what would happen if Trump threatened our allies with ejection from the US security umbrella, a “user fee” on Treasury repayments abroad, or a selective freezing of Treasury repayments altogether, as Miran has suggested in his magnum opus.

Ditto forcing others to “reprofile” into 100-year zero coupon bonds. As the safest and most liquid asset in the world, US Treasury bonds are the bedrock of the global financial system — if they suddenly became less safe and less liquid, a financial panic akin to the Lehman Brothers and coronavirus meltdowns could ensue, taking the US and global economies down with it. The dollar might indeed fall, but not in a way that Trump would like.

All told, a Mar-a-Lago Accord would represent huge downside risk for approximately zero upside gain.

It is doubly amazing that Trump officials seem to be drawn to it when there is another policy that could simultaneously lower the dollar, narrow our trade deficit, reduce interest rates, and put the federal budget on a sustainable path for years to come: cut spending, responsibly raise taxes, and reduce the fiscal deficit.

Instead, we get DOGE, tariffs with a half-life of 1 1/2 hours, threats to our closest allies, and the trashing of America’s credibility. It’s going to be a long four years.

Letter in response to this article:

Mar-a-Lago accord recalls the Thucydides Trap / From Meng Lu, Beijing, China

Apologies for the slow Further Reading today. Here’s a bumper edition to get you through the weekend.

Elsewhere on Friday . . . 

— How the UK tech secretary uses ChatGPT for policy advice (New Scientist)

— Novelist, Nobel laureate and Nazi (The Critic)

— For superfans, comic-con culture is more than fun — it’s sacred (The Conversation)

— Nick Leeson AMA (Reddit)

— On stockpicking in volatile environments (Albert Bridge Capital)

— VAR is a case study of whether fair is actually better (Guardian)

— A ‘new’ Frank Lloyd Wright house has risen in Ohio. But is it legit? (Artnet)

— Who could possibly have predicted this? (WSJ)

— The market power and welfare effects of institutional landlords (Job Market Paper)

— Shayne Coplan’s big bet is paying off (New York)

— Astronomers discover evidence of a stealthy supermassive black hole hiding right ‘under our noses’ (Smithsonian)

Stay informed with free updates

Simply sign up to the Private equity myFT Digest — delivered directly to your inbox.

A year ago self-help guru Tony Robbins released a book called The Holy Grail of Investing, a paean to alternative investments in general, private equity in particular, and private equity firm shares especially.

Lately, it has looked more like when Nazi collaborator Walter Donovan chugged from a fake Holy Grail in the last of the good Indiana Jones films. Here’s a chart showing the relative performance of the larger American alternative asset managers since the turn of the year.

It ain’t Tesla, but it’s definitely not a great look. Some of this is just giving up some of the market-beating gains many of the large alternative asset managers made in recent years, but there does seem to be a more fundamental reassessment of the industry’s outlook.

More traditional asset managers like BlackRock and Franklin Resources are also down, but by far less. So what caused this Robbins-defying private capital quake? Some might suggest there have been some ill omens of late.

For a more rigorous analysis of what ails alternative manager stocks, here’s what Goldman Sachs’ Alex Blostein wrote overnight, with his emphasis below:

Alternative Manager stocks are down 14% YTD (26% off recent peaks and 12% below pre-Election levels) as growth concerns and policy uncertainty drove market volatility higher. The sharp decline has largely been a function of rising earnings risks to Capital Markets sensitive earnings streams (realization income, transaction fees, deployment-based fees, retail flows, etc.) and normalization in the group’s frothy multiples, with 2026 P/E (net of SBC) now at 21X vs. ~24X pre-Elections.

We expect the capital markets environment to remain relatively more challenged over the near-term, pressuring the group’s PRE recovery and to some degree weighing on FRE growth amid risks of slower capital deployment, more elongated capital raising campaigns and moderation in retail flows – resulting in a mid-single digit negative EPS revisions across the group.

That said, we believe the group’s sharp de-rating does not properly reflect the sector’s increasingly durable earnings mix, lower FRE risks than perceived (management fees grew in every prior market downturn), and increasing dividend yield support for several stocks with asset-light balance sheets and high/FRE-covered pay-outs. In the report, we present a bottom-up stress scenario analysis, flexing Performance Related Earnings to below-cyclical averages, slowing fundraising / deployment activity, and reducing level of transaction fees through 2026.

The analysis points to about 15% EPS risk in 2025/2026 on average from a more challenging capital markets backdrop; CG, STEP, and BX derive a relatively higher portion of earnings from capital-markets sensitive sources (in the 30%-40% range), resulting in a 25%- 30% risk to consensus 2025/2026 EPS in our stress scenario. APO and BAM screen as having the least EPS risk in our stress scenario of <10% relative to 2026 consensus. Overlaying these scenarios with stress valuation/recent stock price performance shows OWL, TPG, KKR, and BAM as the most favorable tactical near-term risk reward, while the risk/reward on HLNE, STEP and BX is relatively more balanced.

In other words, Blostein still wants to stay friends with the companies he covers (they are Wall Street’s biggest fee generators, after all) but he reckons that earnings and valuations are coming down across the board due to the current financial and economic turmoil.

You might wonder what all the recent financial and economic uncertainty means for private capital investments. Well, here at Alphaville we aim to also provide timely, forward-looking analysis, so here are our highly-rigorous estimates for probable portfolio marks YTD.