Wim Zwanenburg: NVIDIA, a chipmaker with macroeconomic clout

November 5, 2026

When Nvidia published its quarterly results on 26 August, investors worldwide held their breath. And not just because of the company itself. Nvidia is now regarded as a barometer for the entire AI investment cycle and has therefore taken on almost macroeconomic significance.

Nvidia’s rise has been unprecedented. Between the end of August 2016 and the end of August 2026, Nvidia’s market capitalisation rose from approximately $32.82 billion to $5.33 trillion. That is more than 162 times as much. Over that period, an average annual return of around 66.3 per cent was achieved. With a weighting of 7.7 per cent in the S&P 500, Nvidia now influences not only ‘active’ technology investors – who seek to outperform the index through their overweight or underweight positions – but also broad passive index portfolios. Nvidia’s market capitalisation now approaches Germany’s annual gross domestic product and is approximately four times that of the Netherlands. This comparison is illustrative, but not economically accurate: market capitalisation represents expected future cash flows, whilst GDP measures a country’s annual output.

The latest results highlighted the pace of growth. In the second quarter, which ended in late July, Nvidia recorded revenue of $96.2 billion. This was more than 100 per cent higher than a year earlier and exceeded both market expectations of around $92 to $93 billion and Nvidia’s own forecast of $91 billion. Nvidia’s preliminary forecast of around 70 per cent revenue growth in the 2028 financial year suggests that the AI investment cycle will remain robust for the time being.

Nevertheless, the risks are mounting. Hyperscalers are investing such enormous sums in data centres and GPU chips that some of these cloud providers are running short of free cash flow. They are having to supplement their funding requirements by issuing bonds. At the same time, major clients are developing their own chips to reduce their dependence on Nvidia. Higher memory chip costs, export restrictions to China and competition from Chinese firms could also impact Nvidia’s margins and growth.

If OpenAI, Anthropic and a few hyperscalers continue to account for virtually all demand, a concentration risk arises. Nvidia therefore wishes to broaden its revenue base and, to this end, is supporting the financing of its own ecosystem. The company is providing partial residual value guarantees within a platform of banks and investors on Wall Street that aims to mobilise $500 billion in capital for AI infrastructure. Nvidia is also investing in customers and offering more generous payment terms on major contracts. Critics therefore refer to this as ‘circular financing’: is the end demand really that strong, or is Nvidia artificially creating demand for its own AI products?

Nvidia sometimes supports investments that ordinary lenders do not fully understand or are unwilling to finance, because Nvidia’s management believes it has a much better grasp of the actual demand for AI capacity. This bears a strong resemblance to a central bank that has an information advantage over individual market participants. The key question is whether Nvidia is building up future demand or merely bringing it forward. In the event of widespread, profitable adoption of AI, these financing structures prove to be strategic ecosystem investments that enable Nvidia to win the innovation race. If utilisation rates, rental prices and residual values of AI hardware and data centres fall short of expectations, Nvidia will bear not only the technological market risk but also the financing risk. It is precisely for this reason that Nvidia has become both a microeconomic success story and a macroeconomic test case.
 

Read the full column in Financial Investigator magazine