> However, infrastructure projects tend to have high multipliers - think the interstate highway system and the internet. In theory. In reality that can be the

> However, infrastructure projects tend to have high multipliers - think the interstate highway system and the internet. In theory. In reality that can be the case but is very hard to forecast. Those projects rely on 30+ year estimates of travel habits. Much of the fibre infrastructure put in the ground during the dotcom bubble would still not really be profitable. On the other hand, road demand is far above expectations because of ecommerce and more complex supply chain. Doesn't mean we should do these projects but financing them purely out of debt because of higher anticipated growth is dangerous.