It’s an increase in economic development of a single sector at the expense of the rest of the economy. You’ll notice all the other bars on that graph are showing declines while the data center build-out is booming.
Production is shifting away from the rest of the lagging economy into booming data center development (resource movement effect), and at the same time, all those resources being moved is also triggering movements in labor demand towards providing services to the booming sector (spending effect). We also see USD appreciating and exchange rates going up.
I think that’s one way of putting it but i’m not sure how every other boom couldn’t also be named dutch disease by the same metric
The mechanism behind it is:
The presumed mechanism is that while revenues increase in a growing sector (or inflows of foreign aid), the given economy’s currency becomes stronger (“appreciates”) compared to foreign currencies (manifested in the exchange rate). This results in the country’s other exports becoming more expensive for other countries, while imports become cheaper, rendering those sectors less competitive.
—-
the reason i have to wonder is because this is a graph on internal construction spending and I don’t know what “ai exports” look like nor the currency appreciation because of it
Which this guy also mentions:
The analogy has limits, however. Classic Dutch disease works largely through a stronger currency that undermines exporters, whereas the AI version, as Thompson frames it, operates through competition for scarce inputs. AI could also eventually lift productivity across the wider economy, a benefit a gas discovery never offered. Other forces are at work too, notably energy prices driven higher by the Iran war, which add to electricity costs and inflation independently of AI demand.
Whether the diagnosis holds will depend on data still to come. Electricity prices, wage figures and the pace of AI-linked corporate bond issuance are the measures most likely to show whether the boom is crowding out the rest of the economy or eventually lifting it.
Other booms don’t necessarily cause contraction in the rest of the economy. That’s what the bar graph is showing, it isn’t just showing that there’s a lot of data center construction, it’s showing that construction of everything else is falling.
It definitely might be a new thing, though, because AI doesn’t actually create value. In other resource booms, oil or lithium or whatever, there’s a real material thing being sold for profit. With AI, there aren’t even any profits. It’s just gobbling investment.
It’s an increase in economic development of a single sector at the expense of the rest of the economy. You’ll notice all the other bars on that graph are showing declines while the data center build-out is booming.
Production is shifting away from the rest of the lagging economy into booming data center development (resource movement effect), and at the same time, all those resources being moved is also triggering movements in labor demand towards providing services to the booming sector (spending effect). We also see USD appreciating and exchange rates going up.
It looks a lot like Dutch disease.
I think that’s one way of putting it but i’m not sure how every other boom couldn’t also be named dutch disease by the same metric
The mechanism behind it is:
The presumed mechanism is that while revenues increase in a growing sector (or inflows of foreign aid), the given economy’s currency becomes stronger (“appreciates”) compared to foreign currencies (manifested in the exchange rate). This results in the country’s other exports becoming more expensive for other countries, while imports become cheaper, rendering those sectors less competitive.
—-
the reason i have to wonder is because this is a graph on internal construction spending and I don’t know what “ai exports” look like nor the currency appreciation because of it
Which this guy also mentions:
The analogy has limits, however. Classic Dutch disease works largely through a stronger currency that undermines exporters, whereas the AI version, as Thompson frames it, operates through competition for scarce inputs. AI could also eventually lift productivity across the wider economy, a benefit a gas discovery never offered. Other forces are at work too, notably energy prices driven higher by the Iran war, which add to electricity costs and inflation independently of AI demand.
Whether the diagnosis holds will depend on data still to come. Electricity prices, wage figures and the pace of AI-linked corporate bond issuance are the measures most likely to show whether the boom is crowding out the rest of the economy or eventually lifting it.
https://investinglive.com/news/is-ai-giving-the-us-economy-a-case-of-dutch-disease-from-dutch-gas-to-american-chips/
Maybe have to invent a new term, us disease or something
Other booms don’t necessarily cause contraction in the rest of the economy. That’s what the bar graph is showing, it isn’t just showing that there’s a lot of data center construction, it’s showing that construction of everything else is falling.
It definitely might be a new thing, though, because AI doesn’t actually create value. In other resource booms, oil or lithium or whatever, there’s a real material thing being sold for profit. With AI, there aren’t even any profits. It’s just gobbling investment.
It’s definitely strange.