The world might have entered a period in which economies simply would not grow at the rate they once did. Radical as the thought might seem, it might prove true.
A few years ago, the economist Larry Summers stirred much debate when he suggested that the anemic growth of recent years might not be just a temporary affliction and might have little to do with the 2008 financial crisis.
Instead, he said, it could reflect secular stagnation — a new normal of low consumption and lagging growth stemming from accumulated household debts and rising inequality, among other factors.
Photo: EPA
In different terms, considering the impacts of technological innovations, economist Robert Gordon has been arguing for much the same conclusion.
For two decades now, a lesser-known group of mostly German economists has been making a more extreme argument: standard model of exponential growth — in which an economy can be expected to expand by a given percent every year, no matter how big it gets — is fundamentally flawed.
Rather, these economists claim that while exponential growth fits some young economies, mature economies tend, as a rule, to grow much more slowly — in a linear way, meaning that the percentage growth rate would constantly decline.
The latter view has gained support from a new study, in which a team of European economists and statisticians looked at data on the economic development of 18 mature economies, including the US and most major European nations, from 1960 to 2013 — they started at 1960 to avoid the effects of World War II.
They found that the data on growth in GDP per capita fit best, statistically speaking, with a linear model.
In other words, linear growth — in which mature economies add less new activity in per capita, percentage terms each year — is the empirical norm.
If the finding holds up, then today’s economics might stand in need of some serious conceptual change. As the authors of the new study say, an awful lot of conventional economic analysis rests on the unquestioned assumption of exponential growth.
Governments, for example, rely on it when they decide how much money they need in their social security funds, or when summing up the costs and benefits of any proposed project, including measures to mitigate climate change.
If growth is not exponential, the discounting procedures used habitually in such analyses make no sense at all and standard economics systematically undervalues the future.
The idea of exponential growth rests at the core of essentially all modern theories of growth — theories purporting to explain how capital, labor and technology combine to increase productivity.
Perhaps Summers and Gordon are correct that the fast growth seen over the past couple centuries was a unique, unparalleled episode and that future growth would be much slower. Although the new study does not get into the specific drivers of linear growth, it is consistent with their conclusion.
China’s economic planning agency yesterday outlined details of measures aimed at boosting the economy, but refrained from major spending initiatives. The piecemeal nature of the plans announced yesterday appeared to disappoint investors who were hoping for bolder moves, and the Shanghai Composite Index gave up a 10 percent initial gain as markets reopened after a weeklong holiday to end 4.59 percent higher, while Hong Kong’s Hang Seng Index dived 9.41 percent. Chinese National Development and Reform Commission Chairman Zheng Shanjie (鄭珊潔) said the government would frontload 100 billion yuan (US$14.2 billion) in spending from the government’s budget for next year in addition
Advanced Micro Devices Inc (AMD) suffered its biggest stock decline in more than a month after the company unveiled new artificial intelligence (AI) chips, but did not provide hoped-for information on customers or financial performance. The stock slid 4 percent to US$164.18 on Thursday, the biggest single-day drop since Sept. 3. Shares of the company remain up 11 percent this year. AMD has emerged as the biggest contender to Nvidia Corp in the lucrative market of AI processors. The company’s latest chips would exceed some capabilities of its rival, AMD chief executive officer Lisa Su (蘇姿丰) said at an event hosted by
TECH JUGGERNAUT: TSMC shares have more than doubled since ChatGPT’s launch in late 2022, as demand for cutting-edge artificial intelligence chips remains high Taiwan Semiconductor Manufacturing Co (TSMC, 台積電) yesterday posted a better-than-expected 39 percent rise in quarterly revenue, assuaging concerns that artificial intelligence (AI) hardware spending is beginning to taper off. The main chipmaker for Nvidia Corp and Apple Inc reported third-quarter sales of NT$759.69 billion (US$23.6 billion), compared with the average analyst projection of NT$748 billion. For last month alone, TSMC reported revenue jumped 39.6 percent year-on-year to NT$251.87 billion. Taiwan’s largest company is to disclose its full third-quarter earnings on Thursday next week and update its outlook. Hsinchu-based TSMC produces the cutting-edge chips needed to train AI. The company now makes more
NEXT GENERATION: The new 3-nanometer chip has 28 percent more transistors and offers up to 80 percent faster language model performance than its predecessor MediaTek Inc (聯發科) on Wednesday launched a new flagship smartphone chip, Dimensity 9400, made with Taiwan Semiconductor Manufacturing Co’s (TSMC, 台積電) enhanced 3-nanometer technology, aiming to bring more artificial intelligence (AI) applications to edge devices like phones. The Dimensity 9400 is the second smartphone chip using TSMC’s second-generation 3-nanometer technology, after Apple Inc’s A18 Pro chip for the new iPhone 16 series. The new mobile chip has 28 percent more transistors, offers up to 80 percent faster large language model performance and is up to 35 percent more power-efficient than its predecessor, Dimensity 9300, MediaTek said. Chinese smartphone makers Xiaomi Corp (小米),