“For many, it can be rather intimidating. But anyone who has been able to get through the equity valuation section of the CFA level 2 exam should have absolutely no problem with this.”
The gist of the report is that stocks are pricing in negative growth rates at these valuations. The thing that scares me about the research is that Goldman is using the current 5-year treasury yield of 0.9% as the risk-free rate (I would prefer the 10-year rate using some form of the historical average). Alone, this assumption is pretty dangerous given where we are relative to historical interest rates.
No, I have a framed cartoon of Ben Graham in my office with a quote of “wall street people learn nothing and forget everything” to make sure I am never surprised by garbage like this.
Frankie, have you ever thought of hiding your birth certificate and running for president of the USA? If you ran the US like you ran your personal portfolio, the world would be such a better place.
Don’t most people do this to solve for the implied ERP? Given the numbers I see, I would think that reasonable increases in earnings (say consistent with history) over time would actually be consistent with a very high implied ERP in their model. So rather than showing that earnings are expected to decline, it would rather show that the market is expecting strong growth in equities, even if there is only modest growth in earnings.
you’d think that the ppl working at Goldman would know this, but guess what, they problably think its the right thing to do…i have seen ppl convince themselves of stupidness including myself…but this is quite bad…
I don’t see what the deal is with the focus on the risk-free rate. I mean the 5-year T-note is yielding around 0.9%. The only reasonable way I can think of to improve on that is to use the yield curve (preferably a zero-coupon yield curve) so that there is a different risk-free rate rate in each period. This would ensure the perpetuity at the end has a higher discount rate.
I don’t see 0.9% as the RFR as so unreasonable. The RFR needs to be tuned to your holding period. So if you are an index investor, I’d tend to use something like 2%, which is where the 10y T-note is right now. However, if you have an active strategy, then something like 90d tbills might be better, which are at around 0.1%. The 0.9% rate is somewhere between the 5y and the 7y T-note rate, so is reasonable for people who may not be fully buy-and-hold but don’t intend to trade often, other than rebalancing.
I agree with jmh that I would think that a lot of the issue is less about declining earnings and more about the fact that the Equity Risk Premium is time-varying and is kinda high right now because people are scared. (Incidentally, the flip side of this is that - as long as disaster doesn’t strike - those who aren’t afraid to invest are getting higher rates of return in compensation).
I don’t know, I can’t see the point of using current rates for any type of analysis when you know that we’re in the middle of an unprecedented financial experiment and with luck we’ll never see rates as low as this again.
Like Chad said, you could still factor in current low rates by using a 10 year average or whatever, but to base a valuation decision on the RFR remaining at 0.9% is well…crazy.
The 10 year treasury yield is what the risk free rate is right now (unless you want to use real rates, as STL suggests, which isn’t a bad idea). Right now, if you want to invest your money in 10 years and know how much you are getting back, you need to get a 10y zero coupon T-note. That doesn’t really exist, but a coupon note with small coupons is probably pretty close.
That really is what you are going to get if you aren’t willing to take any risk. If you want more, then you need to take risk.
In fact, that’s part of the point of Fed policy - to keep people investing money in risky assets. The hope is that this will find its way to the buisness sector, where expansion might improve employment. What it really seems to be doing is finding its way to the banks, who use those low rates mostly to try to restore their balance sheets before the next crisis happens.
But since we’re talking about appropriate rates, why limit yourself to one interest rate? Shouldn’t you interpolate the curve and get different points up to 30 years and more? This would solve the question of whether 5y, 10y, or others are the best rate to use; just use different rates for each period. It seems like the simple DDM model is for mathematical convenience rather than accuracy.
Edit: I see that jmh already said something like this, but people just ignored him for the most part…
Yeah jmh has a better solution, but we’re more now talking about the usefulness (or otherwise) of using 0.9%. If somebody brought me some work that used 0.9% I’d tell them to go away and use either a 10 year average and/or what jmh suggests.
Most people take five years of earnings and then choose an exit multiple. At 5 years, the yield curve is still around 1%, which is not so different from 0.9%. Perhaps you’d use a higher rate when determining an appropriate exit multiple, but although I’ve also thought about jmh’s technique, I think it only makes sense when the yield curve is really steep up front. The errors in long term earnings projections are likely to massively outweigh any extra precision you’re getting with the discount rate.