Portfolio evaluation and correcting for the wealth effect

Ask your investment, budget, and other money related questions here
jacob
Site Admin
Posts: 11646
Joined: Fri Jun 28, 2013 8:38 pm
Location: USA, Zone 5b, Koppen Dfa, Elev. 620ft, Walkscore 73
Contact:

Portfolio evaluation and correcting for the wealth effect

Post by jacob »

When indexes and portfolios are setting daily records at nosebleed multiples it's easy to get carried away thinking oneself richer than last year even if the number of shares owned remains largely unchanged and earnings growth remains underwhelming. A nasty side-effect of this is the so-called wealth effect in which one gets tempted to increase spending because one feels richer even as one doesn't own proportionally more assets as well.

I've been fiddling around with my spreadsheets to come up with some more "realistic numbers" that are less influenced by irrational exuberance. To maintain parsimony, I've handled equity, REITs, and bonds in three different but simple ways. I use google sheets and googlefinance() to automatize everything.

Equity:
For each company "X" or index(?) I compute the value as GOOGLEFINANCE(X,"eps")*15.76, where 15.76 is the historical median CAPE10 value that I apply willy-nilly to all companies I own. Each company has its own historical value, but I figure they average out.

REITS:
Here I compute the value as the annual_dividend/0.9*15.76 knowing that REITs are supposed to pay at least 90% of their earnings out as dividends and I multiply by 15.76 again.

Bonds:
I decided to keep it simple. I calculate the market_price*market_yield/0.0384 where the number is the historical median 10year treasury rate. I use the ten year because it's a good average of the entire bond market and because my bond holdings are spread about evenly across the yield curve... and because the multiplier is convenient to find.

Cash:
Cash is cash at face value.

(Of course other numbers are needed for international evaluations including historical currency conversions.)

Results:
The value of my liquid networth is 75% of its current market price.

Seppia
Posts: 1179
Joined: Tue Aug 30, 2016 9:34 am
Location: Italy

Re: Portfolio evaluation and correcting for the wealth effect

Post by Seppia »

25% lower is a surprisingly low number, I thought the “real downside” was much bigger.
I understand markets can go down much more than your “fair value” estimate, but still.
I also understand it is probably influenced by your specific asset allocation, do you know what it would be for a typical 60/40 portfolio?
I’m sorry for asking a question I could probably answer by myself, but I am browsing with a phone and will be for the next 15 days.

I’m probably overly conservative, but I personally always count on about 50% of today’s value for my equities, so my personal portfolio downside would be around 35%

jacob
Site Admin
Posts: 11646
Joined: Fri Jun 28, 2013 8:38 pm
Location: USA, Zone 5b, Koppen Dfa, Elev. 620ft, Walkscore 73
Contact:

Re: Portfolio evaluation and correcting for the wealth effect

Post by jacob »

My number is highly influenced by my particular allocation.

For a balanced 60/40, it would currently be worth 0.6*15.76/24.24+0.4*1.8/3.84=58% of market price.

User avatar
Bankai
Posts: 610
Joined: Fri Jul 25, 2014 5:28 am

Re: Portfolio evaluation and correcting for the wealth effect

Post by Bankai »

Interesting! Should there not be a rule to apply 15,76x only to stocks with PE over that number? Otherwise, one can just buy a basket of low-quality stocks (ie single-digit PE) and conclude one can immediately retire since the 'true' value of one's assets is greater than the current market value.

jacob
Site Admin
Posts: 11646
Joined: Fri Jun 28, 2013 8:38 pm
Location: USA, Zone 5b, Koppen Dfa, Elev. 620ft, Walkscore 73
Contact:

Re: Portfolio evaluation and correcting for the wealth effect

Post by jacob »

Technically the calculation should treat each stock's CAPE10 individually. T's CAPE10 which I'm guessing is around 10ish would be lower than AMZN's which I'm guessing is triple-digit. However, to keep it simple, I just use the average which is valid as long as I hold a mix, that is, my portfolio P/E is not that different from the market's P/E.

Of course one can always invent a metric that allows one to delude oneself in order to build castles in the air but that's what "forward earnings" is for :-P

Quadalupe
Posts: 126
Joined: Fri Jan 23, 2015 4:56 am
Location: the Netherlands

Re: Portfolio evaluation and correcting for the wealth effect

Post by Quadalupe »

Cool! It could also be used for determining asset allocation*.

I was just thinking last week about how one could alter the permanent portfolio by including *some* (possibly wrong) knowledge about 'overvaluedness'. E.g. at a say a P/E of 100/1000/10000, I'm not sure if you'd want 25% in stocks. Jacob's approach gives a nice rule of thumb for bonds/stocks and something similar can be done for gold.

For example, you could say that saner PP allocation right now would be 0.25*15.76/24.24 = ~16% instead of 25% for stocks and 0.25*1.8/3.84 ~= 12% for bonds. So 100-(16+12) ~= 72% should be put into gold/cash (and probably not 36%/36%, but I'm not sure yet on how/where to determine 'under/overvaluedness' of gold yet). In general: the preferred allocation would be 25% * measure_of_overvaluedness, where measure < 1 means overvalued, and measure > 1 means undervalued.

If the stock markets at some points crash to a P/E of 7.88, this formula implicitly recommends upping your stock allocation to 50% (25% * (15.76/7.88). Or maybe it should be renormalized again to 100%, i.e. (50% stocks + 25% gold + 25% bonds + 25% cash normalizes to 40% stocks, 20% gold, 20% bonds, 20% cash)

(* Don't trust me on this, I'm still at the beginning of McConnell's Economics of the ERE curriculum :lol: )

jacob
Site Admin
Posts: 11646
Joined: Fri Jun 28, 2013 8:38 pm
Location: USA, Zone 5b, Koppen Dfa, Elev. 620ft, Walkscore 73
Contact:

Re: Portfolio evaluation and correcting for the wealth effect

Post by jacob »

The historical Dow/Gold ratio is available and is somewhere in the low teens with the current value being in the high teens.
https://www.longtermtrends.net/dow-gold-ratio/

I didn't consider gold above (myopic failure---I don't hold any gold) but if I did, that's how I would go about it.

2Birds1Stone
Posts: 837
Joined: Thu Nov 19, 2015 11:20 am
Location: Earth

Re: Portfolio evaluation and correcting for the wealth effect

Post by 2Birds1Stone »

Very interesting thought exercise! I too discount my equities by 50% in long term planning scenarios.

Now I'm breaking my spreadsheet trying to get the GOOGLEFINANCE function to work -_-

5ts
Posts: 26
Joined: Sat Nov 30, 2019 2:43 pm

Re: Portfolio evaluation and correcting for the wealth effect

Post by 5ts »

I appreciate this approach because I think we are feeling way too good about ourselves because of the last 10 years. This had led us to flights of fancy and unrealistic withdrawal rates and overzealous assumptions about our wealth. I wish ERE on everyone, but let's be realistic about this.

IlliniDave
Posts: 2675
Joined: Wed Apr 02, 2014 7:46 pm

Re: Portfolio evaluation and correcting for the wealth effect

Post by IlliniDave »

For the last several years, longer than I've been hanging here, I've done something similar and estimate what I call a "cash out net worth". I'm not as rigorous about calculating the equity (over)valuation discount, and I include estimates of tax hits. Given the amount I have in pretax, mine's < 75% by a good bit, more like 50% or maybe 55% now that I'm longer on cash/bonds. And that might be a tad optimistic. Probably should revisit the arithmetic for that one of these days.

I also couple my estimate of future returns to current yields and valuations. I just use SP valuation as a proxy for all stocks and assume that by the time I turn 70 valuations will be back down to my eyeball average of post-1973 valuations (~22 IIRC). Interestingly, despite the recent/current benchmark records, Shiller PE was generally lower in 2019 than it was in 2017 and 2018, so my computed future outlook is slightly rosier now which is contrary to what my gut tells me. Basically though, I'm still expecting < 2.5% pa real return on a 60/40 portfolio over the next 15 years.

black_son_of_gray
Posts: 381
Joined: Fri Jan 02, 2015 7:39 pm

Re: Portfolio evaluation and correcting for the wealth effect

Post by black_son_of_gray »

Quadalupe wrote:
Sat Jan 04, 2020 11:47 am
Cool! It could also be used for determining asset allocation*.
...Ta-da! Hussman has a white paper (his own introduction, the white paper .pdf) on strategic allocation based on relative valuation in a mixed portfolio. Some of the implementation (i.e. his calculation of market "internals") is a little opaque/proprietary, but much of it is fairly basic valuation math. The core ideas are well worth consideration.

classical_Liberal
Posts: 1186
Joined: Sun Mar 20, 2016 6:05 am

Re: Portfolio evaluation and correcting for the wealth effect

Post by classical_Liberal »

I think this exercise is fascinating, even if I disagree with some of the underlying mechanics @jacob is using. My biggest beef(s) in this realm is probably with REIT's, those PE's are generally higher due to dividend payouts. Also, cash can be susceptible to inflation, maybe it should be discounted a bit based on low rates/low inflationary thought process default from recency bias. Lastly, and most importantly, all of these asset classes rarely move together, even if they are all inflated, capital has to go somewhere.

It's important not to take a pessimistic double dip when playing this game though. I realize there are plenty of doomers here, but if we are trying to compare today's values to historical norms, there should be some consistency. If someone adjusts their Net Worth down based on current valuations to historical norms AND they assume lower than average returns for the asset class going forward, me thinks this exercise could get out of control. This is fine for those running at 1% WR rates already, if it makes them feel more secure. However, it could also keep some people working way too long. IOW, if someone is already assuming a mix portfolio only yielding 1-3% real in the next ten years, asset values have already been accounted for to a large extent.

Question. If one believes that they need to discount all these assets up to 50% due to valuations, what justification is their to not sell? The only one I can think of is taxes, but this can probably be dealt with pretty efficiently with medium term plans, particularly given the current low tax rates in place.

IlliniDave
Posts: 2675
Joined: Wed Apr 02, 2014 7:46 pm

Re: Portfolio evaluation and correcting for the wealth effect

Post by IlliniDave »

classical_Liberal wrote:
Sat Jan 04, 2020 3:46 pm

It's important not to take a pessimistic double dip when playing this game though. I realize there are plenty of doomers here, but if we are trying to compare today's values to historical norms, there should be some consistency. If someone adjusts their Net Worth down based on current valuations to historical norms AND they assume lower than average returns for the asset class going forward, me thinks this exercise could get out of control.
Yes, this is a good point. When I apply my relatively low forward-looking returns, it gets applied to current market value, not the "cash out" number. Overall my "cash out net worth" is more of a curio than something I pay attention to. It started when I didn't have a lot invested and was looking at how long I could get by if I literally sold everything and lived off the proceeds. It's good to have an understanding how volatile valuations and taxes would affect you should you have to raise cash for ransom money or something, but I think the expectation is usually a more orderly liquidation and to my way of thinking peanut-buttering an unwinding of high PEs over a period of time is a pretty good "down the middle" estimate.
classical_Liberal wrote:
Sat Jan 04, 2020 3:46 pm
Question. If one believes that they need to discount all these assets up to 50% due to valuations, what justification is their to not sell? The only one I can think of is taxes, but this can probably be dealt with pretty efficiently with medium term plans, particularly given the current low tax rates in place.
Can't speak for jacob, but for me the discounting is just a look at a highly unlikely scenario (the need to raise a ton of cash quickly). For me net worth (standard calculation or "custom" calculation) is interesting but not a terribly meaningful number for planning purposes. If you assume an instantaneous near-term PE decline is a certainly maybe you would want to cash out, otherwise it's just a lower bound resulting from a low probability event.
Last edited by IlliniDave on Sat Jan 04, 2020 5:27 pm, edited 1 time in total.

Lucky C
Posts: 379
Joined: Sat Apr 16, 2016 6:09 am

Re: Portfolio evaluation and correcting for the wealth effect

Post by Lucky C »

If you have access to a longer earnings history, longer PE# (# = number of years) has higher correlation to longer subsequent time periods of returns, at least for the broad stock market. E.g. PE30 vs. 20-year subsequent returns has a much higher correlation than PE10 vs. 10-year returns.
https://greenbackd.com/tag/shiller-pe/

Of course longer periods would further distort old mature value companies vs. new growth companies (some of which haven't been around 20 or 30 years), so you'd have to restrict yourself to looking at only "old" stocks or the broad market. Perhaps Jacob's stock-by-stock method can be combined with a PE20 or PE30 of the whole market vs. its long term average. This will get you out of trouble with stocks that have had an unusually high profit margin streak the past decade looking cheap (see also Hussman's MAPE).

5ts
Posts: 26
Joined: Sat Nov 30, 2019 2:43 pm

Re: Portfolio evaluation and correcting for the wealth effect

Post by 5ts »

classical_Liberal wrote:
Sat Jan 04, 2020 3:46 pm
Question. If one believes that they need to discount all these assets up to 50% due to valuations, what justification is their to not sell? The only one I can think of is taxes, but this can probably be dealt with pretty efficiently with medium term plans, particularly given the current low tax rates in place.
There is no point in selling. Not all assets are overvalued but at least some are. If you agree with this then you tilt things in your favor with value stocks or funds and then build in a fudge factor for the inevitable overvalued things you didn't identify.

If you don't agree, then you pick your asset valuation, within reason, and go from there. Since we are on the ERE forum, I'm sure you will be fine regardless of how things work out, since you have multiple income streams and multiple skills you can parlay into work should things go upside down or sideways.

Laura Ingalls
Posts: 324
Joined: Mon Jun 25, 2012 3:13 am

Re: Portfolio evaluation and correcting for the wealth effect

Post by Laura Ingalls »

Isn’t the sequence of returns the real enemy as opposed to the wealth effect?

We “declared” ourselves FI ~six years ago. The market was at all time highs then. Net worth has increased ~65% since then. I don’t know the timing and depth of the next bear cycle but I am a realist and know it will happen sooner or later. But I also know that an almost 40% pull back to be back to 2014 levels.

If 40% less money was enough then it should be enough now, as I am six years closer to my eventual death regardless of when it ultimately occurs. The same logic applies to the financial aspects of child rearing. Two kids with an average age of 16,5 should require fewer financial resources than two kids with an average age of 10.5.

ertyu
Posts: 292
Joined: Sun Nov 13, 2016 2:31 am

Re: Portfolio evaluation and correcting for the wealth effect

Post by ertyu »

Laura: The wealth effect is, you think you're richer than you are, so you spend more than you can actually afford because your brokerage account balance says you can. Thus correcting for the wealth effect would be, "the number here might say 100k, but in actuality, it is 58k and I should spend accordingly. I imagine that's the point of this exercise.

Other than that, yes: it is sequence of returns/initial amount

IlliniDave
Posts: 2675
Joined: Wed Apr 02, 2014 7:46 pm

Re: Portfolio evaluation and correcting for the wealth effect

Post by IlliniDave »

I was thinking about this a little bit overnight. I think the main reason I maintain a set of sorta conservative metrics is really to create a shock absorber for myself--if I wake up one morning with a 50% haircut I've at least played through the scenario mentally. Besides anticipating muted returns and looking at substantially reduced initial balances I have a collection of scenarios that I can toggle through, some of which will heap additional misery onto the pile. I'm kind of a Jurassic specimen in the sense that I'm expecting a modest little retirement annuity and being US-based, SS, so in a future that plays out on the benign side I'm only partly dependent on the stash for a window of time, at least for basic lifestyle maintenance.

A few days ago in a different thread I was discussing what amounts to the wealth effect with some folks. In 2019 I was guilty of pulling forward some optional lifestyle expenses (optional, but important to me). It's quite possible that subconsciously I had fallen prey to the wealth effect. In the front of my mind it was a matter of an opportunity appearing that I had prepared for. The spending was relatively minor, on the order of 0.5% of financial assets at market value, but it wouldn't be hard to create a hypothetical scenario where that 0.5% would matter. An alternate view is that by evoking the fungibility of money principle, and the fact that it would take a near-doomsday scenario for the 0.5% to matter, and that barring that doomsday the purchases were going to be made anyway; I'd basically engaged in a little market-timing and "sold high" to fund the purchases. That rationale is a little hollow-sounding to my ears, but it's not completely ridiculous.

Laura Ingalls mentioned sequence of returns, and that's one I don't play explicitly through iDaveSim. I don't know what the science would say but I have a strong intuition that SoR risk is higher when one starts retirement at higher valuations. My way of mitigating it is to borrow from Bernstein and Pfau and using a liability matching scheme: I maintain an estimate of lifetime withdrawals and plan to keep at least 2X that amount in relatively stable value assets (versus 1X for standard liability matching). That number isn't really affected by instantaneous portfolio balance.

ertyu
Posts: 292
Joined: Sun Nov 13, 2016 2:31 am

Re: Portfolio evaluation and correcting for the wealth effect

Post by ertyu »

what are "relatively stable value assets" these days tho

Seppia
Posts: 1179
Joined: Tue Aug 30, 2016 9:34 am
Location: Italy

Re: Portfolio evaluation and correcting for the wealth effect

Post by Seppia »

classical_Liberal wrote:
Sat Jan 04, 2020 3:46 pm
Question. If one believes that they need to discount all these assets up to 50% due to valuations, what justification is their to not sell?
As for Dave, I don’t discount my equities 50% because I believe that is their correct value.
It’s what I consider a realistic “worst case scenario”.
Visualizing this eventuality I believe would help me deal with it should it materialize.

The main reason not to sell, for me, is always the same:
Markets are so unpredictable in the short term that being completely out of the market can be extremely damaging.
I don’t know about others, but I personally considered markets to be significantly overvalued as far as in 2014.
Imagine if I had cashed out back then.
When one is in the accumulation phase, I think the best thing to do in case of big fears of a crash is to simply stop adding new positions.
Ere people saving >50% will relatively quickly build up significant dry powder without giving up the potential upside.

Post Reply