by Hubert Marleau, Market Economist, Palos Management
February 10, 2023
Traders should have been happy with this week's reprieve of top-tier economic data. Unfortunately, they were busy interpreting a cacophony of Fedspeaks, which led to a sharp increase in bond yields pushing the S&P 500 down 46 points or 1.1% to 4090. Fourth Quarter S&P 500 earnings per share are uneven but decreased only 2.9% from the same period a year earlier compared to a year-over-year 12% decline for the broad benchmark. It may account for the reason why 4Q22 EPS misses are not heavily punished. “Of the 345 companies that have reported so far, overall earnings results are beating estimates by a median of 6%, and 69% of those reporting are beating estimates” says Tom Lee.
On Monday, Atlanta Fed President Raphael Bostic said: “ Data like the January NFP report will probably mean we have to do a little more work.” I would expect that would translate into raising interest rates higher than previously forecast, or leaving them higher than originally expected. He reiterated that his base case remains for rates to reach the terminal rate of 5.125%., in line with the median of policymakers’ December forecasts, and stay there throughout 2024.
On Tuesday, Jerome Powell spoke at the Economic Club of Washington zigzagging between harsh and dovish rhetoric, messaging that the focus will be on monthly data and emphasising that 2023 will be a year of significant inflation decline. Tiff Macklem, the Governor of the Bank of Canada (BoC) also spoke. He said: “Inflation is turning the corner and that monetary policy is working, making it unnecessary to raise interest rates further if, as expected, the economy stalled and inflation came down.” The minutes of his Governing Council showed that the Canadian central bank had a more dovish tilt than generally acknowledged.
On Wednesday, John Williams, the head of the powerful NY Fed, including Governors Lisa Cook and Chritopher Waller, gave the market a reality check, pointing out to investors that, if financial conditions were to become too loose, the Fed might be obliged to raise interest rates more than predicted. Minneapolis Fed President Neel Kashkari thankfully got lost in the translation, expecting a higher terminal rate (5.375%) than the official one (5.125%).
Credit conditions have not correlated as they should with the Fed’s monetary stance. Powell has attributed this rare phenomenon to a simple difference of opinion between the markets (stock, bond and forex) and the monetary authorities. This gap however, has narrowed considerably over the past week.
On Thursday, Richmond Fed President Thomas Barkin said: “I’m confident that our foot is unequivocally on the brakes and it just makes sense to steer more deliberately as we study the impact of existing rate hikes.” In this connection, he believes that it will take another 6 to 12 months for demand pullbacks to quiet the rate of inflation.
The Main Economic Prints of the Week
On the inflation front, there was not much this week. The Federal Reserve Bank of New York showed that the global supply chain pressures index decreased in January to 1.00 from 4.50 a year ago. Other than the pandemic period, the index has ranged between -1.00 and +1.00, implying that the supply chain is where it ought to be. Energy supply lines are now basically restructured as well, and therefore prices for natural gas and oil will remain normalised. Meanwhile, the University of Michigan reported that 5-to-10 year inflation expectation remains at 2.9% for the third straight month. The bigger picture is that they have been stuck in a narrow range of 2.7-to-3.1% for over two years. The pre-Covid trend was 2.5%. The January consumer price index is set to be released onTuesday. It could have major implications on how the Fed will react if expectations are met. Street consensus is +0.50%. The Cleveland Fed’s fast moving inflation tracker sees the January CPI climbing by 6.5% y/y and 0.65% m/m. It will be tricky to interpret the new data because various components like shelter and trucks were reweighted by the Bureau of Labour (BLS). It explains why the swap market has shifted the y/y inflation in one year’s time from 2.1% to 2.4% % and 5-year inflation expectation rose from 2.25%% to 2.45% in a matter of days. Price increases are on there way down, but not in a linear fashion.
On the growth front, economic growth should moderate to a slower gear from hereon because a lot of excess savings have been eaten up by soaring prices as pandemic relief has unwound. Jobless claims rose last week for the first time in 6 weeks, and consumers have cut back on their spending as the consumer credit data suggests. Indeed, the stock of credit outstanding rose by only $11.6 billion in December and the personal savings rate grew from a September low of 2.4% to 3.4% in December. Goldman Sachs estimates that the monthly savings rate will rise modestly to 4.5% by year-end because precautionary demand for cash and liquidity will increase. Banks have reported in a recent survey of senior loan officers, that borrowers have reduced demand for loans across a wide range of business and consumer credit fronts Going into 2023, banks are also expected to tighten lending standards even more, because loan quality has deteriorated.
The NowCasting model of the Atlanta Fed is predicting an annual rate of increase in R-GDP of 2.2% in Q1. Should the deflator for personal consumption expenditures rise at the annual rate of 2.1%, as it did in the second half of 2022, the economy would have a total nominal increase of about 4.3%. What we have is an “Immaculate Disinflation” at best, or a “Soft Landing” at worst, but neither a “Hard Recession” nor an “Impending Boom.” What we do know is that the yield curve is predicting an imminent reduction in the nominal increase in GDP because it has been inverted for a long time and the inversion is historically steep.
The real question is whether it will have more of an impact on inflation than growth. Based on a year-over-year comparison inflation is more elevated than growth and broad disinflation is building. I suspect inflation. Why? The yield curve is deepening, while long-dated bond yields are rising as are real rates. This tells me that nominal growth will fall, but not necessarily real growth. Re-acceleration signs like auto sales and housing are undeniable. So what gives? Inflation. Goldman Sachs has a different view here than me. The discrepancy results from investors’ strong belief that the neutral rate - a rate that balances the economy between inflation and employment correctly - is the one that prevailed in the pre-pandemic era (1.75%), accounting for the deep inversion of the yield curve. I don't buy it. The neutral rate today is 3.93%. Admittedly, it is presumptuous to take on Goldman Sachs, but thank god that I have JPMorgan on my side. In a note published to their clients, it said that the yield on long-term bonds might indeed be pricing in a sharp decline in inflation and nothing else.
The New Fed’s Timetable Under High-Employment Disinflation: As I See It
The Fed will likely retreat from hawkish rhetoric in May when the next 3 rounds of growth and inflation data will give assurance that the path of inflation is clearly down as growth moderates either into a very soft landing or specifically into “immaculate disinflation”. This suggests that the Fed will not need to raise interest rates any higher than its current forecast, but keep them higher for longer. Blowout employment numbers convince me that the monetary authorities may not cut interest rates in the latter months of 2023 as I originally believed. Nonetheless investors should look past the current narrative and try to figure out where the puck will be 6 months from now. The economy is not about to return to another period of monetary utopia, but the time-line to an even keel monetary policy is foreseeable. There will be 2 more small rate increases, 1 in March and another in May, because policymakers are wedded to the idea of posturing in front of the market and the economy. Credit conditions are good and the recession risk has receded. Nevertheless, the Fed will likely announce a pause in May to be followed by a long wait before pivoting in Q1/24. That will be accompanied by a formal declaration that the mission is accomplished, as wage growth may have cooled down further to a level compatible with the Fed’s 2% inflation target.
Despite the blowout job reports of last Friday, the Employment Cost Index, Average Hourly Earnings and Unit Labour Costs are all pointing to a slowdown in wage growth. I suspect that the Fed may want by then to steer monetary policy toward a balanced approach. This could happen sooner because the lags appear to be much shorter than previously thought. Darion Perkins at Lombard TS offered 3 explanations. The central banks are transparent, signalling their intentions far into the future, using their balance sheet to support their interest rate policy and the credit system has migrated to freer capital markets. Additionally, the tightening cycle has been significantly more dramatic than in the past. Real rates have moved way up, the gap between the policy rate and neutral rate has corrected and the money supply has actually fallen over the past year. It makes sense to pause and see if the ongoing monetary policy will do what it's supposed to do, as Tiff Macklem has suggested: force inflation down. I don't think that the U.S. monetary authorities will take any cues from the BoC, but surely unanimity among the members of the FOMC is probably fraying at this point. Alan Blinder, former Vice-Chairman of the Federal Reserve, said it well: “The Fed’s job is to reduce inflation, not drive the economy into the ground.” This is especially true when the goal line is close.
The Chatbot Craze: Delivering the Future
We’ve all heard that the rate of technological change is accelerating. The World Economic Forum says it's unprecedented and therefore unpredictable, and with so many new technologies emerging on so many fronts, it's challenging to keep up. The McKinsey Global Institute has compiled a list of 11 such technologies that will transform the global economy by 2025 - an industrial revolution that will alter business and society, potentially disrupting the status quo. These are: Mobile Internet, Artificial Intelligence, Virtual Reality, Cloud Technology, Internet of Things, Advanced Robotics, Biometric Technology, 3D Printing, Genomics, Blockchain and Quantum Computing.
ChatGPT is short for “Chat Generative Pre-trained Transformer”. It is an artificially intelligent creator of content. It is a bot and its technology is an amazing mind-blowing generative AI tool. This software machine can leverage the skills and expertise of people to another level. Tyler Cowen has an embracing definition: “ChatGPT is a conversational language model developed by OpenAI. It is a deep learning-based model trained on a large dataset of text, with the aim of generating human-like responses to questions. ChatGPT is part of the GPT series of models and is based on transformer architecture, which has been the state-of-the-art in many NLP tasks. It has the ability to generate text in a wide range of styles and topics and can answer questions, summarise long documents,generate creative writing, and perform other language tasks.” One can query it with details. In fact the more complex the question, the better.
Approximately 100 million people used ChatGPT in the month of January, which is the fastest-growing user base ever, including buzzy TikTok, according to Bloomberg. Students are already using ChatGPT to write essays; businesses to create websites, promote materials and respond to customer-service inquiries; lawyers to produce legal briefs and academics to generate footnotes. Ordinary people can do many things with it like organise notes, correct grammar, and work with mathematical symbols. AI is also flexible. It can do a better job than Excel, be more creative than Google Doc, and eliminate the search bar. The WSJ’s Joanna wrote that she uses ChatGPT to generate ideas for interviews, emails, columns and video scripts. “This is going to help us do our jobs better and reduce some of the drudgery. It is a productivity boost.
The speed of its adoption suggests that its progress may be more expansionary than linear and, therefore, threaten white-collar jobs, as this new ‘technological bot” replaces skilled office workers whose jobs were once considered immune to automation. An Oxford study estimates that 47% of U.S. jobs might be at risk. As a matter of fact, AI systems get better as they get more used, absorbing more data and in turn eliminating updates. In an interview with Barro’s, BigBear CEO Mandy Long said that she thinks that ChatGPT has been used inside government and commercial applications for close to 2 decades. Stories abound. No wonder, then, that market is enthusiastic about this space. Companies like BigBear.ai, C3.ai and SoundHound are the talk of the town. There are a bunch of startups like AI21 Labs and Anthropic that are working on natural language-processing models like ChatGPT, that help computers understand the way humans write and speak. In this context one should watch Stability AI, AnyWord, BuzzFeed, CloserCopy, JasperAI, Frase.io, WriteSonic, and CopySmith. They all generate content similar to ChatGPT.
In the meantime, it’s too much hype for me because there is still a lot of work needed here to make this technology a necessary tool for everybody. Nevertheless, investors should not discount the possibility that generative AI is the next big thing. It should be taken seriously as being more than just a fad because it could be a creative destruction moment like the telegraph, railroads, radio, automobiles, dotcoms, computers, and smartphones. All those innovations changed how we live and work, but the guys who really made big money were the ones who piled in early. Unfortunately it is not easy to enter the game, even for venture-capital players. Morgan Stanley calculated a few days ago that, collectively the top 10 private generative AI companies are valued at $30 billion. The horse has bolted. In this regard, I’d rather stick with the biggest long-term winners, where generative AI is their top priority: Microsoft, Google, IBM, Adobe, Nvidia, Advanced Micro Devices and perhaps Apple. (After all, the Apple phone is an AI-automated machine tool.) Incidentally, Generative AI needs huge amounts of computing power to operate efficiently. It could bring a major shakeup in the semiconductor and cloud computing industries.
Copyright © Palos Management
