Arnold Kling raised an interesting question: The Puzzle of Low Real Interest Rates and Low Investment. He summarises the various possibilities into Keynesian and non-Keynesian:
The puzzle in macroeconomic data is that the real interest rate is low and investment is low. There are a number of stories, none of them fully convincing.
In the Keynesian category, we have:
1. Low “animal spirits.” As far as I know, no one has actually propounded this.
2. Accelerator model. That is, when other forms of spending are high, investment is high. So when spending by households goes down, investment goes down. I put Furman (and most Keynesians) in this camp.
In the non-Keynesian category, we have:
3. Real interest rates are actually high, because prices are falling. This is perhaps more plausible if you think about sectoral price movements. If the price indexes go up because of college tuition and health insurance, then prices elsewhere may be falling.
4. Real interest rates are actually high for risky investment. Interest rates on government debt and on high-grade private bonds are a misleading indicator of the marginal cost of capital.
5. Crowding-out can occur at low interest rates. That is, if financial intermediaries are gorging on government debt, they may not seek out private-sector borrowers.
The puzzle in my view is a result of thinking entirely in money and not looking at the real side of the story at the same time. See Chapters 16 and 17 of my Free Market Economics, but essentially the reason for low money rates creating a fall in the real level of investment is because the supply of savings diminishes, and then, because the price of these savings to those who can get the money is so low, many really low productivity and no productivity activities ended up being funded. The analysis may be faulty somehow, but the conclusions it reaches is what you see in the world around us today.