Important security note: Warning of attempted fraud in the name of DWS
We have detected that fraudulent individuals are misusing the "DWS" trademark and the names of DWS employees on the internet and social media. These fraudsters are operating fake websites, Facebook pages, WhatsApp groups and Mobile Apps. Please be aware that DWS does not have any Facebook Ambassador profiles or WhatsApp chats. If you receive any unexpected calls, messages, or emails claiming to be from DWS, exercise caution and do not make any payments or disclose personal information. We encourage you to report any suspicious activity to info@dws.com, including any relevant documents and the original fraudulent email. Additionally, if you believe you have been a victim of fraud, please notify your local authorities and take steps to protect yourself.
U.S. unemployment steady, nerves frayed
05/12/2025
A widening gap between official data and worker sentiment points to the power of technological anxiety.
Transitions driven by technology or trade are rarely painless. Will artificial intelligence (AI) be any different? At first glance, the U.S. unemployment rate—holding steady at 4.4%—should reassure. Look closer, and the picture is more ambiguous. Our Chart of the Week tells two stories. Unemployment has barely budged since 2021. Job growth has slowed since the 2024 presidential election, but most indicators suggest a gradual, not an abrupt, deterioration. Yet, with tighter immigration controls, unemployment rates reveal less than usual. Simultaneous shifts in trade and migration policy, implemented with little warning, risk structural mismatches and widening sectoral wage gaps.
Curiously, these sectoral divergences are not yet the main story. Instead, working-age Americans’ expectations appear more pessimistic. The second curve in our chart tracks net expectations of losing one’s job within 12 months, versus expectations of being employed again within 3 months, if the current job was lost today. Historically, this measure of workers’ expectations has mirrored the unemployment rate. Now, however, workers appear increasingly anxious about their prospects, even as headline joblessness remains low.
This hints, in our view, at something deeper. AI may be the buzzword of the year, but so far, evidence of its economic impact on typical U.S. workplaces—let alone aggregate productivity—remains elusive.1 By contrast, uncertainty about AI, rather than the algorithms themselves, may already be shaping sentiment. The mere prospect of technological change—amplified by high-profile layoffs and persistent macroeconomic uncertainty—has fuelled a sense of fragility. The recent U.S. government shutdown, which delayed September’s jobs report (now published), only added to the fog. While the delayed figures showed little change in unemployment, the subjective measures of worker anxiety have continued to climb.
That could have implications for policy. As economic historian and recent Nobel Prize winning economist Joel Mokyr noted three decades ago, “Technological progress reduces the wealth of those possessing capital (real or human) specific to the old technology that cannot readily be converted to the new.”2 Often, it is expectations of what new technologies might do—not their actual impact—that drive behaviour.
“It’s not the numbers that keep people up at night—it’s the uncertainty about what comes next,” argues Christian Scherrmann, U.S. Economist at DWS. In 2026 and beyond, the real impact of AI may be less about the jobs it replaces today, and more about the uncertainty it sows for tomorrow. The gap between perception and reality is itself a force that can shape responses of policymakers and firms alike.
Job separation expectations are rising, even as measured unemployment has largely held steady
Sources: Federal Reserve Bank of New York, Haver Analytics, DWS Investment GmbH as of 11/30/25
* Net expectation: expectation of job loss within the next 12 months minus the confidence of finding a new job within 3 months (if job loss occurred today)
This information is subject to change at any time, based upon economic, market and other considerations and should not be construed as a recommendation. Past performance is not indicative of future returns. Forecasts are based on assumptions, estimates, opinions and hypothetical models that may prove to be incorrect. Alternative investments may be speculative and involve significant risks including illiquidity, heightened potential for loss and lack of transparency. Alternatives are not suitable for all clients.