Will artificial intelligence have a dampening effect on inflation and interest rates, and a positive effect on productivity and profit margins? Let’s just say we shouldn’t count on it saving the day.
The past 30 years witnessed the advent of globalization, the admission of China to the World Trade Organization, cheap labour, lower taxes, weakening labour union negotiating power and increased global trade. Inflation and interest rates declined and corporate profit margins rose sharply. Profit margins for S&P 500 companies increased from 6 per cent in 2000 to about 10 per cent in 2019. Profit margins and stock prices are highly correlated. Increased price-to-earnings multiples also contributed to rising valuations.
But globalization is over. Wars tend to lead to an end of globalization, and Russia and China are both challenging it. The events in Ukraine are auguring a new Cold War and will present other difficulties, and protectionism and nationalism are on the rise, with industries moving back to their home, higher-cost countries, as seen with the United States and Japan.
The surge of new manufacturing in North America is increasing the demand for labour. The U.S. lost 20 million jobs in 2020 owing to COVID-19 and the retirement of baby boomers who are replaced by smaller numbers of less-productive younger workers. As a result, union strike activity has risen sharply and so have wages, not only in the U.S., but in Canada and Britain. Unions are regaining their negotiating power.
Taxes are also on the way up. Pandemic-related government deficits and ballooning debt will require higher taxes going forward. Total outstanding U.S. federal debt stands now at US$33-trillion. Fourteen per cent of federal tax revenues are now spent to service that debt. This will lead to higher taxes.
Underinvestment in the oil and gas sector and in mining owing to heavy regulation and the ESG (environmental, social and governance) craze have shifted the price dynamics for all commodities. The resulting shortages will lead to higher prices in the face of rising demand due to electric vehicle production and renewable energy.
As a result, inflation will rise, secularly, and profit margins will be eroded going forward. Nominal interest rates will go up, not only because inflation expectations will rise, but also because real interest rates will be on the way up, secularly.
In the long run, real interest rates will rise. Demographics are one force pushing them higher. Baby boomers have been retiring and have stopped saving; in fact, they are in their decumulation years, which reduces the supply of funds. This is happening in the face of increased demand for capital by corporations that need to embed innovation and new technologies into their production processes, as well as by governments that need to borrow to fund structural deficits.
To clear the supply-demand imbalance, real interest rates are pushed up, like what happened in the late 1970s. Moreover, while globalization added to global liquidity, deglobalization will have the opposite effect.
The impact of higher interest rates, along with the erosion of profit margins, will be predictably negative, in contrast to the positive effect that lower interest rates and higher profit margins had on stocks over the past 30 years. The outlook appears particularly unfavourable for growth stocks, which would be hurt the most by higher inflation and interest rates.
AI may surcharge productivity, but for now it is not clear exactly how AI will render business processes more effective.
Historically, new technologies have been disappointing in terms of increasing productivity. Despite technological advances since the 1980s, productivity has fallen by 50 per cent in the U.S. and Japan. Robert Solow, the Nobel Prize winner, has written that the computer age showed up everywhere except in productivity stats.
At this point, it is not clear at all what AI’s effect on reducing the need for labour and on labour costs will be. For example, self checkouts have failed to live up to their promise. According to recent reports, they have not led to more efficient checkouts, and have increased store thefts. Moreover, there is little to suggest that they have reduced the need for workers.
And computer chipmaking jobs are highly specialized. The U.S. wants to build its supply of qualified workers, but it will take time – ask Intel Corp., which cannot find enough qualified people to work in the new U.S. plant it is building. Eventually, some workers will be replaced by automation, but the remaining highly specialized jobs will continue to command high wages.
Finally, AI may have lost some of its original lustre. Traffic volume on Chat GPT has been falling. Job postings for AI specialists are now 50-per-cent below what they were last year. And the chief executive of Nvidia Corp. has been selling shares in his company. In two weeks this past September, he sold US$70-million worth of stock.
More specifically, though, AI may lower inflation by reducing the number of workers and/or by making workers more productive. But at the same time, it will also increase demand for capital, which will put upward pressure on real interest rates, and offset or negate the decline in inflation.
No matter how one looks at the inflation and interest-rate problems, AI won’t alleviate them. Value investing will beat growth, riding the exchange-traded fund bandwagon will not work, and stock picking and fundamental analysis will be key to outperformance going forward.
George Athanassakos is a professor of finance and holds the Ben Graham Chair in Value Investing at the Ivey Business School, University of Western Ontario.
Be smart with your money. Get the latest investing insights delivered right to your inbox three times a week, with the Globe Investor newsletter. Sign up today.