Raphael Stevens

All 23 of the U.S. banks included in the Federal Reserve’s annual stress test weathered a severe recession scenario while continuing to lend to consumers and corporations.
The rate of total loan losses varied considerably across the banks, from a low of 1.3% at Charles Schwab to 14.7% at Capital One; credit cards were easily the most problematic loan product.
Banks including JPMorgan Chase and Wells Fargo are expected to disclose updated plans for buybacks and dividends Friday after the close of regular trading.

Michael Barr, Vice Chair for Supervision at the Federal Reserve, testifies about recent bank failures during a US Senate Committee on Banking, House and Urban Affairs hearing on Capitol Hill in Washington, DC, May 18, 2023.
Saul Loeb | AFP | Getty Images

All 23 of the U.S. banks included in the Federal Reserve’s annual stress test weathered a severe recession scenario while continuing to lend to consumers and corporations, the regulator said Wednesday.
The banks were able to maintain minimum capital levels, despite $541 billion in projected losses for the group, while continuing to provide credit to the economy in the hypothetical recession, the Fed said in a release.

Begun in the aftermath of the 2008 financial crisis, which was caused in part by irresponsible banks, the Fed’s annual stress test dictates how much capital the industry can return to shareholders via buybacks and dividends. In this year’s exam, the banks underwent a “severe global recession” with unemployment surging to 10%, a 40% decline in commercial real estate values and a 38% drop in housing prices.
Banks are the focus of heightened scrutiny in the weeks following the collapse of three midsized banks earlier this year. But smaller banks avoid the Fed’s test entirely. The test examines giants including JPMorgan Chase and Wells Fargo, international banks with large U.S. operations, and the biggest regional players including PNC and Truist.
As a result, clearing the stress test hurdle isn’t the “all clear” signal its been in previous years. Still expected in coming months are increased regulations on regional banks because of the recent failures, as well as tighter international standards likely to boost capital requirements for the country’s largest banks.  
“Today’s results confirm that the banking system remains strong and resilient,” Michael Barr, vice chair for supervision at the Fed, said in the release. “At the same time, this stress test is only one way to measure that strength. We should remain humble about how risks can arise and continue our work to ensure that banks are resilient to a range of economic scenarios, market shocks, and other stresses.”

Goldman’s credit card losses

Losses on loans made up 78% of the $541 billion in projected losses, with most of the rest coming from trading losses at Wall Street firms, the Fed said. The rate of total loan losses varied considerably across the banks, from a low of 1.3% at Charles Schwab to 14.7% at Capital One.

Credit cards were easily the most problematic loan product in the exam. The average loss rate for cards in the group was 17.4%; the next-worst average loss rate was for commercial real estate loans at 8.8%.
Among card lenders, Goldman Sachs’ portfolio posted a nearly 25% loss rate in the hypothetical downturn — the highest for any single loan category across the 23 banks— followed by Capital One’s 22% rate. Mounting losses in Goldman’s consumer division in recent years, driven by provisioning for credit-card loans, forced CEO David Solomon to pivot away from his retail banking strategy.

Regional banks pinched?

The group saw their total capital levels drop from 12.4% to 10.1% during the hypothetical recession. But that average obscured larger hits to capital — which provides a cushion for loan losses — seen at banks that have greater exposure to commercial real estate and credit-card loans.
Regional banks including U.S. Bank, Truist, Citizens, M&T and card-centric Capital One had the lowest stressed capital levels in the exam, hovering between 6% and 8%. While still above current standards, those relatively low levels could be a factor if coming regulation forces the industry to hold higher levels of capital.
Big banks generally performed better than regional and card-centric firms, Jefferies analyst Ken Usdin wrote Wednesday in a research note. Capital One, Citigroup, Citizens and Truist could see the biggest increases in required capital buffers after the exam, he wrote.
Banks are expected to disclose updated plans for buybacks and dividends Friday after the close of regular trading. Given uncertainties about upcoming regulation and the risks of an actual recession arriving in the next year, analysts have said banks are likely to be relatively conservative with their capital plans. More

A banner for the online image board Pinterest Inc. hangs from the New York Stock Exchange on the morning Pinterest made its initial public offering, April 18, 2019.
Spencer Platt | Getty Images News | Getty Images

Check out the companies making the biggest moves midday.
Pinterest — Shares climbed 6.59%. Wells Fargo upgraded Pinterest to overweight due to an Amazon partnership expected to take hold later this year and optimism that Pinterest can continue to boost user engagement.

Cruise stocks — Carnival popped 8.81%, Norwegian Cruise Line gained 7.55% and Royal Caribbean added 1.68%, extending gains from Tuesday after Carnival reported a smaller-than-expected loss for its second quarter and issued strong guidance. The sector has been on a tear this year as it recovers from the Covid-19 pandemic.
General Mills — Shares tumbled 5.17% after the maker of Betty Crocker mixes and Cheerios cereal turned in a mixed earnings report for its fiscal fourth quarter. The company exceeded Wall Street expectations on earnings, posting $1.12 in adjusted earnings per share against a consensus estimate of $1.07 from analysts polled by Refinitiv. But $5.03 billion in revenue missed analysts’ forecast of $5.17 billion.
Chip stocks — Shares of Nvidia slipped 1.81% and Advanced Micro Devices was down 0.2%, paring earlier losses, following a Wall Street Journal report that the U.S. is weighing new restrictions on artificial intelligence chip stocks sold to China.
Netflix — The streaming giant jumped 3.06% after Oppenheimer raised its price target to $500 per share from $450. The Wall Street firm said it anticipated more subscribers and the potential discontinuation of its lowest-priced, ad-free plan, which is being tested in Canada.
Joby Aviation — Shares soared 40.22% after the company announced it received a permit to begin flight testing its first electric vertical takeoff and landing vehicle (eVTOL).

AeroVironment — Shares added 4.86% after the military drone maker reported revenue of $186 million after the market close Tuesday, topping analysts’ projection of $164 million, according to consensus estimates from Refinitiv. AeroVironment also said it anticipates full-year revenue of $630 million to $660 million, beating the $600 million expected by analysts.
ZoomInfo — The software stock rose 6.09% after Needham initiated coverage of ZoomInfo with a buy rating. Needham said in a note to clients that ZoomInfo has “best in class unit economics.” ZoomInfo also received positive coverage from Morgan Stanley, which reiterated an overweight rating on the stock.
Snowflake — Shares added 3.86% after the data cloud company reiterated its full-year guidance during an investor day Tuesday. Goldman Sachs reiterated its buy rating on Snowflake after the event and Morgan Stanley maintained an overweight recommendation.
Circor International — The maker of flow control products for industrial and aerospace and defense markets users rallied 4.25% following a Reuters report that private equity firm Arcline has offered $57 per share, topping a rival bid from KKR.
First Citizens BancShares — The regional bank gained 0.4%. Atlantic Equities initiated coverage of the North Carolina bank Wednesday with an overweight rating and $1,775 per share price target, which suggests nearly 50% upside from Tuesday’s close.
— CNBC’s Alex Harring, Brian Evans, Jesse Pound and Michael Bloom contributed reporting. More

A gradual reverse migration is under way, from Zoom to the conference room. Wall Street firms have been among the most forceful in summoning workers to their offices, but in recent months even many tech titans—Apple, Google, Meta and more—have demanded staff show up to the office at least three days a week. For work-from-home believers, it looks like the revenge of corporate curmudgeons. Didn’t a spate of studies during the covid-19 pandemic demonstrate that remote work was often more productive than toiling in the office? Unfortunately for the believers, new research mostly runs counter to this, showing that offices, for all their flaws, remain essential. A good starting point is a working paper that received much attention when it was published in 2020 by Natalia Emanuel and Emma Harrington, then both doctoral students at Harvard University. They found an 8% increase in the number of calls handled per hour by employees of an online retailer that had shifted from offices to homes. Far less noticed was a revised version of their paper, published in May by the Federal Reserve Bank of New York. The boost to efficiency had instead become a 4% decline.The researchers had not made a mistake. Rather, they received more precise data, including detailed work schedules. Not only did employees answer fewer calls when remote, the quality of their interactions suffered. They put customers on hold for longer. More also phoned back, an indication of unresolved problems.The revision comes hot on the tails of other studies that have reached similar conclusions. David Atkin and Antoinette Schoar, both of the Massachusetts Institute of Technology, and Sumit Shinde of the University of California, Los Angeles, randomly assigned data-entry workers in India to labour either from home or the office. Those working at home were 18% less productive than their peers in the office. Michael Gibbs of the University of Chicago and Friederike Mengel and Christoph Siemroth, both of the University of Essex, found a productivity shortfall, relative to prior in-office performance, of as much as 19% for the remote employees of a large Asian it firm. Another study determined that even chess professionals play less well in online matches than face-to-face tilts. Yet another used a laboratory experiment to show that video conferences inhibit creative thinking.The reasons for the findings will probably not surprise anyone who has spent much of the past few years working from a dining-room table. It is harder for people to collaborate from home. Workers in the Fed study spoke of missing their “neighbours to turn to for assistance”. Other researchers who looked at the communication records of nearly 62,000 employees at Microsoft observed that professional networks within the company become more static and isolated. Teleconferencing is a pale imitation of in-the-flesh meetings: researchers at Harvard Business School, for example, concluded that “virtual water coolers”—rolled out by many companies during the pandemic—often encroached on crowded schedules with limited benefits. To use the terminology of Ronald Coase, an economist who focused on the structure of companies, all these problems represent an increase in co-ordination costs, making collective enterprise more unwieldy.Some of the co-ordination costs of remote work might reasonably be expected to fall as people get used to it. Since 2020, many will have become adept at using Zoom, Webex, Teams or Slack. But another cost may rise over time: the underdevelopment of human capital. In a study of software engineers published in April, Drs Emanuel and Harrington, along with Amanda Pallais, also of Harvard, found that feedback exchanged between colleagues dropped sharply after the move to remote work. Drs Atkin, Schoar and Shinde documented a relative decline in learning for workers at home. Those in offices picked up skills more quickly.The origins of the view that, contrary to the above, remote working boosts productivity can be traced to an experiment nearly a decade before the pandemic, which was reported by Nicholas Bloom of Stanford and others in 2013. Call-centre workers for a Chinese online travel agency now known as Trip.com increased their performance by 13% when remote—a figure that continues to appear in media coverage today. But two big wrinkles are often neglected: first, more than two-thirds of the improved performance came from employees working longer hours, not more efficiently; second, the Chinese firm eventually halted remote work because off-site employees struggled to get promoted. In 2022 Dr Bloom revisited Trip.com, this time to investigate the effects of a hybrid-working trial. The outcomes of this experiment were less striking: it had a negligible impact on productivity, though workers put in longer days and wrote more code when in the office.The price of happinessThere is more to work (and life) than productivity. Perhaps the greatest virtue of remote work is that it leads to happier employees. People spend less time commuting, which from their vantage-point might feel like an increase in productivity, even if conventional measures fail to detect it. They can more easily fit in school pick-ups and doctor appointments, not to mention the occasional lie-in or mid-morning jog. And some tasks—notably, those requiring unbroken concentration for long periods—can often be done more smoothly from home than in open-plan offices. All this explains why so many workers have become so office-shy.Indeed, multiple surveys have found employees are willing to accept pay cuts for the option of working from home. Having satisfied employees on slightly lower pay, in turn, might be a good deal for corporate managers. For many people, then, the future of work will remain hybrid. Nevertheless, the balance of the work week is likely to tilt back to the office and away from home—not because bosses are sadomasochists with a kink for rush-hour traffic, but because better productivity lies in that direction. ■ More

Getting into Jinjiang Ode is a little difficult. The luxurious property development in central Chengdu will not allow potential buyers through its four-metre-high palatial gates without an appointment. Even finding out about the project in the south-western metropolis, home to 16m people, is tricky. The firm behind it is so confident of demand that it does not deign to advertise the flats—a confidence which is not unjustified. Chengdu has a distinct, laid-back atmosphere epitomised by its public tea gardens, in which patrons spend hours sipping hot beverages and having their ears cleaned. The leisurely pace of life and tongue-numbing local cuisine appeal to younger Chinese people, who have come in droves in recent years, says Zhang Xiaojun, a sales agent at the development. Many of them buy homes.As a prolonged downturn in China’s property market takes hold, Chengdu seems to be an outlier. By several metrics, including house prices and sales of new homes, it is faring better than almost anywhere in the country. At a national level, the central government’s response to the deepening property crisis, including an interest-rate cut announced on June 13th, has underwhelmed. China’s benchmark stock index has fallen by 8% since its peak this year in early May, when the country still appeared to be rocketing towards a full post-covid recovery. Now investors fear more developers will start to fall short of cash, defaulting on dollar debts in the process. Experts are asking how much local measures can pump up growth. Chengdu is a good place to search for answers.There is a faint air of unreality about the local market. New home sales between April and June were 30% higher than in the same period in 2019, the last year before the covid-19 pandemic struck, notes Larry Hu of Macquarie, an investment bank. In contrast, across China’s 30 largest cities, sales have fallen by 25%. Meanwhile, in May home prices in Chengdu rose by 8% compared with the previous year, the most of any large city. It has notched month-on-month rises for 17 straight months. Many Chinese cities are working through vast inventories of flats that have been built but not sold: it will take the southern city of Zhuhai more than 12 years to sell homes that have been completed or are still under construction if sales stay at the current pace. Chengdu will sell such flats in just over three years.What explains the success? Since 2016 officials in every Chinese city have been able to devise their own measures for cooling or heating local property markets. Most of the rules employed are restrictions on who can buy a flat, how many they may purchase and the size of the downpayment required. In most large cities, only people with local hukou, or residence permits, are allowed to buy homes. In Chengdu, high-level purchase controls remain in place. But officials have sought to attract families as a way of expanding the city and increasing demand for homes. Residents with two or more children are, for instance, allowed to buy additional homes, and local hukou-holders may buy up to three. Even those without a hukou may buy two. Since the start of the year, elderly parents who move to Chengdu to join their adult children may also purchase a flat.Other cities have experimented with similar policies, but enjoyed much less success. Shenzhen, the technology hub across the border from Hong Kong, has relaxed some of its restrictions. Yet property prices are still down 1.8% year-on-year. One explanation for this is sweeping layoffs in the city’s tech sector. Another is that Chengdu’s policies are more effective because they are paired with reforms to attract educated workers, which have helped boost growth. Since 2017 local authorities have handed out housing subsidies and cash rewards to talented people who move to the city in order to work in its rapidly growing industrial base, points out Sandra Chow of CreditSights, a research firm.Chengdu’s officials also did a better job of tackling the crisis of confidence that spread across the country last year. As developers went bust, many failed to finish flats. Thousands of homebuyers responded by halting mortgage payments. Many more delayed buying new homes. Officials in Chengdu went to great lengths to ensure homes were handed over, funnelling cash to developers, says Ms Chow. Even defaulting developers managed to complete homes. About 40% more apartment space was finished in the first two months of 2023 compared with the same period the year before. This probably encouraged wavering buyers to take the plunge. Other regions may have wanted to follow suit, but lacked the cash. Sichuan, where Chengdu sits, notched up the strongest growth in municipal land sales of any province in the first half of 2022, which will have freed up funds to keep builders at work.Chengdu benefited from some other factors that will be difficult, if not impossible, to replicate elsewhere, and perhaps even in the city itself. Its population rose by more than 7m from 2011 to 2021, making it one of the fastest-growing urban areas anywhere in the world. These inflows have been the biggest driver for housing demand, says Yan Yuejin of E-House China, a research firm. But urban migration has since slowed. There are simply not enough people in China for another population boom. Chengdu’s location in the south-west also meant it did not see rapid rises in prices in past housing booms. Moreover, its growing manufacturing industry continued to lift incomes. As Louise Loo of Oxford Economics, another research firm, notes, it is thus one of just a few second-tier cities that have not seen rapid price increases relative to local incomes.A few levers remain for Chengdu’s officials should things start to look peaky. They have yet to drastically ease restrictions, allowing many more people to buy homes. Market-watchers are waiting for such a development, says Guo Jie of the Local Association of Real Estate Enterprises, an industry group, for it would indicate that steam is running out and that even the best-prepared cities are being swept into the crisis. Policymakers elsewhere in the country will be watching closely, too. ■ More