Caught in the crosscurrents of progress and peril, the global economy today is at the crossroads with the separation between fiscal and monetary policies – long a cornerstone of central bank autonomy and credibility – coming under growing strain.
This is the nub of the Annual Economic Report of the world’s central bankers’ central bank, the venerable Bank for International Settlements (BIS) released at Basel on June 28 in its annual general meeting.
Without mincing words, it said that despite successive shocks from tariffs to the conflict in West Asia, the global economy evinced and evidenced “surprising resilience”. This was partly driven by optimism around progress in artificial intelligence (AI) that fuelled massive AI-related investments, growing financial vulnerabilities and weakening fiscal positions in major economies.
But the euphoria over AI and the financial system vulnerabilities in exposing itself is real and rattling, it warned adding that Big Tech’s AI outlay binges risk tapering off in a prolonged “investment bust” that could shake financial system and weaken the world economy. The prospect of worse than expected returns in the tech sector could compel investors to swiftly curb financing for AI firms at a juncture when the five biggest ‘hyperscalers’ are likely to invest more than $1 trillion from 2025 to the end of this year. While laying out the potential risks of the AI exuberance, the BIS pointblank noted that “disappointments in returns could trigger a sudden pullback in financing and turn the capital expenditure (capex) boom into a protracted investment bust, with potential knock-on effects on financial conditions’ globally.
The bugle of caution comes amid emerging concerns over the scale of equity and debt issuance stoking the AI frenzy and the turmoil it is triggering in global markets. Tech companies have zeroed in on the global credit markets in recent months, raising hundreds of billions of dollars to fund AI projects, taking due advantage of corporate credit spreads that are at a nadir over the recent decades. A reversal of AI optimism could, the BIS said, “have major financial consequences, given AI firms’ rising leverage and burgeoning footprint in credit markets”.
Already, experts have warned that AI progress in automated tasks could fail to translate into faster aggregate growth If essential human performed tasks hold output back. This is particularly so if different economic tasks are strong complements, the overall output would be stymied by the task that is improving slowest, the weakest link, the report cited Trammell and Korinek (2023) who also contend that economies currently operate in this complementarity regime, implying that the transformative scenario remains “a possibility rather than a present reality”.
Tellingly the report highlighted how the opacity of AI sector financing has compounded the vulnerabilities. Hyperscalers, chipmakers and AI labs are linked through a complex web of private pacts with the visible one being circular financing: chipmakers and hyperscalers take equity stakes in AI labs or neo cloud providers, who in turn commit to multi-year purchases of chips or computing power. Data centre construction is increasingly outsourced to third parties that lease facilities back to hyperscalers on long-dated contracts with embedded exit clauses. The terms of such deals are poorly disclosed with risks of the same assets being pledged multiple times. Together, such pacts account for a sizable share of sector-wise financing and forward revenue!
An interesting facet documented in the report is the ongoing AI boom posing challenges for monetary policy, especially for emerging market economies (EMEs) including India and small open economies (SOEs) where it is shaped by a shifting external milieu. Thus, capital flows could become more selective and volatile as geoeconomic fragmentation redraws the cross-border investment landscape along strategic as well as financial lines, leaving them more exposed to abrupt and less predictable capital flow reversals as happened last year in India on a larger scale. It called upon them to follow a flexible inflation-targeting tack with judicious use of foreign exchange intervention, complemented by deeper domestic capital markets and prudent reserve buffers, over and above a credible medium-term fiscal framework.
The AI build-out, BIS said, has recently been facing growing bottlenecks in electricity, advanced semiconductors, and grid equipment. Fast-growing demand for computing power is already pressuring electricity prices and input costs, with potential spillovers to inflation. These interim shortages might also balloon over- investment with firms trying to lock in future capacity through long-dated contracts that further lay them open to any disappointment in demand.
Apart from AI, the report focused on three factors to explain the muted effect of tariffs turmoil triggered by Trump in 2025. Foremost is their smaller than expected size. Because of exemptions, trade agreements and measured responses from trading partners, the effective average US tariff rate stabilised at 10 per cent in the second half of 2025 and accounting for this pulls down global output loss by about a third from initial estimates. Second trade realignment mitigated the impact of tariffs as in the case of China, the sharp decline in US-bound exports was compensated by higher exports to other parts of Asia. Third firms adapted their trading strategies with significant frontloading of trade before tariffs took effect, which cushioned both exporters and importers.
(G Srinivasan is a senior economic journalist based in New Delhi.)
