In the world of economics, there is an indicator considered the "ultimate answer" for measuring the true driving force of an economy's long-term growth: Total Factor Productivity (TFP). This reflects whether an economy can continuously create more production capacity through technological progress and efficiency improvements, given constant capital and labor inputs. Economists believe that TFP almost determines a country's long-term fate. Nobel laureate economist Paul Krugman once pointed out that a country's ability to improve its standard of living over time depends almost entirely on its ability to increase per capita output. And technological progress is at the heart of all this. This article is excerpted from The Blockworks, with key points compiled and edited.
Maintaining a stable US debt-to-GDP ratio ensures financial stability.
The importance of Total Productive Factors (TPF) is not only reflected in abstract growth theories, but also directly relates to the sustainability of government finances. A recent study by the National Bureau of Economic Research (NBER) indicates that if the US government can maintain a stable debt-to-GDP ratio, it only needs an additional 0.5 percentage points of total productive factor growth each year to stabilize US finances.
0.5% may seem insignificant, but its impact is far-reaching. According to the study, if such productivity growth can be sustained for ten years, the baseline forecast for US government debt will decrease by about $2 trillion; looking at 30 years, the debt-to-GDP ratio will be 42 percentage points lower than the baseline forecast, and even 80 percentage points lower than the pessimistic scenario.
Artificial intelligence assisting humans can boost overall productivity
Against the backdrop of record-breaking US fiscal deficits and debt, this outcome seems almost unbelievable. However, researchers at the artificial intelligence company Anthropic believe the technology's potential may extend far beyond this. Anthropic recently analyzed approximately 100,000 real-world conversations using Claude.ai, attempting to estimate the difference in time required for humans to complete the same tasks with and without AI assistance, and further inferring its impact on overall economic productivity. The study concluded that AI assistance has the potential to increase total productivity by approximately 1.1 percentage points.
This figure has attracted widespread attention. If a 0.5% increase in productivity is enough to stabilize government finances for decades, then a 1.1% increase could theoretically have a disruptive impact on the economy and public finances. Of course, academics and policymakers are not without reservations about such an optimistic estimate. Anthropic's research itself acknowledges that its analysis is highly dependent on model assumptions. For example, the study shows that Claude can complete a course design in 11 minutes, saving teachers about 4 hours of work time, but whether "saving time" necessarily translates into "increasing output" remains highly uncertain.
Critics point out that the time saved may not be invested in higher-value economic activities, but could instead be used for entertainment or consumption, such as browsing social media or reading briefings. In this case, while artificial intelligence does improve people's welfare and leisure time, it may not increase overall wealth, and its help to governments in solving debt problems is relatively limited. However, Anthropic also believes that their estimates may be conservative. The study did not take into account the accelerating rate of AI adoption, nor did it consider the additional productivity gains brought about by the continued evolution of future model capabilities. In other words, the study assumes that humans will continue to use language models in the current way and at the current level for the next ten years. Considering that large language models show significant progress almost every few months, and that humans are still rapidly learning how to use them, Anthropic believes that 1.1% may only be an "approximate lower bound" for the productivity effect of artificial intelligence.
More importantly, this study only measures the impact of artificial intelligence on "speeding up the completion of existing tasks," without taking into account the fundamental restructuring of workflows and production methods caused by technology. Anthropic points out that major leaps in productivity throughout history, from electricity and computers to the internet, have not simply made old things faster, but have fundamentally changed how things are done.
Such structural changes are difficult to model, yet often have the most profound impacts. Even so, researchers remain cautious, detailing methodological limitations and assumptions. They also acknowledge that even if artificial intelligence does create greater fiscal space for governments, future legislators could still accumulate debt again through increased spending. However, given the widespread perception of impending fiscal risks, this optimistic scenario, even if only partially materializes, is worth anticipating.
The article "Can Artificial Intelligence Save US Finances?" by Anthropic Analysis points out that AI can improve TFP (Total Productive Factors) and stabilize finances. This article first appeared on ABMedia ABMedia .
