Hacker News new | ask | show | jobs
by boppo1 1362 days ago
Someone correct me if I'm wrong[3]:

1. Buying bonds pushes the price of bonds up (pretty basic, right?)

2. Bonds have a constant interest payment, so if you buy a higher priced bond, the size of that return as a % of what you paid (yield) is smaller.

3. When you evaluate risky[0] securities (company stocks and bonds) or other risky financial investments, you compare them with something risk-free[1] like government bonds.

4. More specifically, to determine how much something is worth now, you calculate Net Present Value (NPV) using the Discounted Cash Flow (DCF) method. This is the method that all serious financial companies employ[2] on some level to determine the prices of things.

5. NPV is the sum of each (future cash flow / (1 + discount rate)^t) where t=compounding periods (typically years) and the discount rate is the benchmark rate that you're comparing to.

6. Hence, if you are using government bond yields as the discount rate in your calculation, if the yield goes down, your NPV goes up. So by lowering government bond yields, you help support the price of financial assets.

7. If you are not familiar with finance, you may be inclined to ask "hold on that seems like it supports stock prices not the actual economy" "what about the value of money?" "what about the quality of activity being supported?" "once you start, how do you stop?". Interfering with the 'natural' rate of interest on benchmark securities was and remains a hotly contested topic, however it has been accepted as a status quo method in western economies that feature a central bank.

[0] not risky as in Gamestop, but risky as in 'any investment that could possibly go badly'

[1] We generally treat US treasuries, and sovereign debt for stable countries like the UK and Germany as 'risk-free'. If these borrowers ever default, we probably have bigger problems to deal with than the valuation of Coca-Cola.

[2] Lots of valuations are not calculated using DCF, companies often just use comparables and 'multiples' (literally just value something by saying it's "10x earnings/ebitda (current period cash flow)". However, this is mostly just companies that need to make lots of valuation calls very quickly and off-the-cuff. Generally speaking DCF is the 'fundamental theory' of valuation, and then other more or less complex models are built off of its ideas. I.E. adding coefficients of probabilities, what-if scenarios, etc to future cash flows.

[3] which is possible given I'm laughably unemployed