Can cash be your friend?

[OutOfTheBlue]

When someone deposits (essentially lends) money to a bank, what is the actual yield the bank gets on that money against the measly 0.4% (or whatever) they offer them in return? Do banks share a fair percentage of their revenue with depositors?

And when something goes wrong, who is being bailed out? That business model does ring some alarm bells too.

I don't want to derail this discussion and dive into more specifics here, but it is not just cash vs bonds and stocks.

At any rate, due diligence is a must (best to invest in something one understands).

And of course, there are other ways to invest in crypto (like staking from a private wallet), I just brought up earning yield on stablecoins as an alternative to keeping (too much) cash around.

[WFJ]

Thanks WFJ
The trinity study was done in 1998, and the average dividend yield on the S&P500 over that timeframe was ~4%. So for most of that time you could just live on dividends from the index. 20 year bond yields were also above 4%. Right now the dividend yield on the S&P500 is 1.5% and bond yields are also much lower than historically. I would argue that at the current point in time you need at least double the amount of money to live of the S&P500/bonds as the trinity study would suggest, so we are at a 2% WR if you want to be safe.
The alternative is to look for better investment strategies. (And no, any portfolio that has only index funds or bond funds in them probably cant work for this, because the underlying expected returns wont change) (and pls dont mention crypto or gold now, inflation adjusted expected returns for these assets are also zero in the long run)

Bingo, using data outside that of the sample period would be called "Extrapolation" and again, anyone involved in stats would completely throw out any analysis outside the sample period dataset. These are the rules. The other issue identified in the post is that of sensitivity analysis and sensitivity to initial condition, which are not robustly tested. I would speculate that all WR estimates are extremely sensitive to real interest rates at the beginning and duration of a retirement journey. Had the 4% rule been framed as "(5 yr note yield + S&P 500 Dividend Yield)/2" this would average around 4% for the sample period, but now be well below 2% and IMHO provide a better estimate of a safe WR for FIRE today. This is not popular and why there is such blowback, despite the mathematically robust arguments.

BUT using simple stats and logical arguments often results in anger and odd responses from the anonymous internet. This is also why I do not like working in front line financial services as dealing with this kind of stuff from clients and even corporate mouthpieces is exhausting in real life. Feel some responsibility to point these shortcomings out as the cult like allegiance to some of the FIRE beliefs has been shocking to me.

[steveo73]

This. I agree being pre-RE (accumulation phase, semiERE, what have you) and being post-RE makes for an inherently different approach in that regard.

Having cash or at least some stable asset on hand in troubled times can give some peace of mind, and that could be priceless!

The point is that lots of assets can have their place. Prior to retirement I didn't like cash or bonds. Now I like seeing those assets in my portfolio.

[steveo73]

When you know already that you are in the tail end event, is it smart to still ignore it?

The point is that you don't know. I think good investing requires humility. You won't know if you are in a tail event until 30-50 years time. That tail event is exceptionally hard to pick.

Imagine owning a house in the midwest. Is it a tail event that your house gets destroyed by a tornado? I am pretty sure it is.
But what if you can see the tornado already on the horizon and its heading to your house, is it still an edge case?
Does it matter at that point?

If you are in a tail event it may matter but these examples are extremely different. You cannot predict the future 30-50 years ahead. You can't. I can't. No one can.

Here is the kicker. WR's are completely opaque. You cannot accurately predict your expenses that far ahead either. You just can't. No one can.

There is a cost as well. So you can't predict the tail event. You have no proof and you won't know until a long long time ahead. Then you are using inaccurate parameters because your expenses are not set in stone. The cost doesn't change though - you have to work longer and assuming you have some fancy portfolio you are taking on more risk.

It's cool to do whatever you want but there are pros and cons. It's not right and wrong. There is nothing wrong with a 60/40 stock bonds portfolio completely in index funds. Plus if you take that option you have really good data to assess your decision. If you don't like that portfolio choose another one.

[steveo73]

When someone deposits (essentially lends) money to a bank, what is the actual yield the bank gets on that money against the measly 0.4% (or whatever) they offer them in return? Do banks share a fair percentage of their revenue with depositors?

And when something goes wrong, who is being bailed out? That business model does ring some alarm bells too.

It is just relatively the safest most liquid asset class. It's not going to save you in a world war.

I don't want to derail this discussion and dive into more specifics here, but it is not just cash vs bonds and stocks.

These lines are for investment newbies (not having a go at you). It's not about just those 3 asset classes and it's definitely not about comparing what is better out of those 3 asset classes or other asset classes.

Portfolio creation is a lot broader than this.

One point that I think a lot of people don't understand with all the criticism against MPT is that you are doing it anyway. Some people just do it based on feelings and they don't have empirical evidence to back up their position. I don't understand though the criticism of MPT. It's non-sensical.

If you invest in crypto and a customized stock portfolio along with futures positions you are still creating a portfolio. You might have no idea about MPT. You might have no idea about backtesting data. You are still creating a portfolio.

[steveo73]

Bingo, using data outside that of the sample period would be called "Extrapolation" and again, anyone involved in stats would completely throw out any analysis outside the sample period dataset. These are the rules.

What ! So the rules state you can't just make up your own data and run your test against your own data.

Sure you can just extrapolate data but what sort of a fool would do that ?

You set up a straw man and you attack that straw man.

If you have better data use it. Please provide that data to myself as well. You'd be able to sell that data as well.

The other issue identified in the post is that of sensitivity analysis and sensitivity to initial condition, which are not robustly tested. I would speculate that all WR estimates are extremely sensitive to real interest rates at the beginning and duration of a retirement journey. Had the 4% rule been framed as "(5 yr note yield + S&P 500 Dividend Yield)/2" this would average around 4% for the sample period, but now be well below 2% and IMHO provide a better estimate of a safe WR for FIRE today. This is not popular and why there is such blowback, despite the mathematically robust arguments.

Dude - you keep doing this,. You need to provide facts and empirical evidence. You are again arguing against a position that no one takes.

Good faith arguments were mentioned earlier in this thread. You are not doing this.

Have you heard of SORR ? Have you read ERN's data analysis ? Your points have been covered in detail by smart people.

Feel some responsibility to point these shortcomings out as the cult like allegiance to some of the FIRE beliefs has been shocking to me.

Let's be extremely clear that yet again you have created some silly straw man to argue against. If there is a cult it's clear the cult is an ERE cult to belittle MPT and using empirical data to base your decisions on which is just stupid. It's like stating well let's ignore the facts because I don't like the facts. The data doesn't matter. It's arrogant and stupid. It's toxic and if creates a toxic environment.

There is no cult like allegiance to FIRE beliefs. There is a thing called MPT. There is a historical data-set to run tests against. If you can provide better data then do it. If not then stop making up straw man arguments that no one is stating.

Lastly - you need to start discussing issues in good faith. You are just arguing against stupid straw men that you've made up. You are not adding anything of any intelligence or nuance to the discussion.

It's wrong and it goes against all the principles people are rallying against on this thread.

@AxelHeyst @Campitor. Posters who have made comments regarding how to post should be responding to these types of posts. Integrity requires having principles and sticking to those principles.

[AxelHeyst]

@AxelHeyst @Campitor. Posters who have made comments regarding how to post should be responding to these types of posts.

No. There's not going to be a posse of ERE schollmarms rapping knuckles. I commented to you (after a years long campaign of trying to ignore you and hope you'd go away) because I find your posts to be consistently and predictably unacceptable, and I wanted to make it clear that I find it so, because Im now afraid that if your behavior is tolerated any longer it'll signal to others with similar blindnesses that it's fine to post here.

Broadly speaking, what I see happening here is that every time someone says something negative about indexing, you infer a personal attack on your intelligence, and respond defensively. You exhibit zero initiative to understand other people's points of views, and consistently interpret posts in the dimmest possible light.

Again, I wouldn't say anything if I just didn't like the way you post. But your posting style does damage to the health of this forum. I've read enough old threads to know that attempting to break down and explain the nuances of this is certain to be a waste of everyone's time, so I've reached the limit of what I can actively do here, I think.

I'll just leave this here.
The City and the City.

[prudentelo]

Had the 4% rule been framed as "(5 yr note yield + S&P 500 Dividend Yield)/2" this would average around 4% for the sample period, but now be well below 2% and IMHO provide a better estimate of a safe WR for FIRE today.

If that is the right way to see it then other implication is salaries halved since then. If you spend most income on FIRE anyway which many here do.

$4,000 net in 1998 and 1JAFI (say $600) spending means $3,400 being spent on FIRE worth $1,700 today for total $2,300 income equivalent today. Ignoring ordinary inflation


Makes me ask why work today. If you spend it all on FIRE. Even if CV gap or whatever halves income possibility between now and mean reversion, still only breaking even. Not losing.

Other side of coin of asset overvaluation is you can turn them into goods and services at favorable pr8ce.

[M]

If that is the right way to see it then other implication is salaries halved since then. If you spend most income on FIRE anyway which many here do.

$4,000 net in 1998 and 1JAFI (say $600) spending means $3,400 being spent on FIRE worth $1,700 today for total $2,300 income equivalent today. Ignoring ordinary inflation


Makes me ask why work today. If you spend it all on FIRE. Even if CV gap or whatever halves income possibility between now and mean reversion, still only breaking even. Not losing.

Other side of coin of asset overvaluation is you can turn them into goods and services at favorable pr8ce.

This makes me wonder, assuming one is in the US and has enough credits to qualify for social security, why not just put everything into tips/I bonds and wait for social security? Is this not the 'risk free' path?

[prudentelo]

Yes but if you graduated in 22 you can retire earliest 32 and still need 30 year expenses saved.

So low risk but expensive.

If youre "late retiree" at like 45 then why not if you believe TIPS inflation matching promise.

[steveo73]

No. There's not going to be a posse of ERE schollmarms rapping knuckles.

Cool. I think the poor behavior is pretty obvious within this thread but that is your call.

[steveo73]

If that is the right way to see it then other implication is salaries halved since then

It's not the right way to view the situation.

This makes me wonder, assuming one is in the US and has enough credits to qualify for social security, why not just put everything into tips/I bonds and wait for social security? Is this not the 'risk free' path?

Your no 1 risk to your portfolio longevity is inflation. The best hedge against inflation is stocks. The assets that can be killed via inflation are cash related assets like bonds and cash.

So a portfolio that is 100% cash related assets has a significant chance of failure and poor real returns over the longer term.

Cash/bonds -> your short term hedge against asset deflation (stocks/property etc).
Stocks -> your long term hedge against inflation.

If you want to have a 100% cash related asset type portfolio you'd better be prepared for the cost of that portfolio which will mean getting a very low WR which means additional years working.

I have a fantastic suggestion for when people have these ideas to discuss. It should be a process:-

1. State your premise
2. Back up your premise with empirical evidence.
3. State the pros and cons of this approach.

=> So in this case the premise is stocks and bonds are overvalued and therefore we are at a tail end of the curve and therefore you require 100% cash related assets.

The evidence for being at that part of the curve is not currently available. You won't have that evidence until we get to the future point which is your focus. It could be 20 years but it could be 50 years or it could be 5 years.

The pros of this approach are that your portfolio is relatively safe over shorter time frames but relatively unsafe over longer time periods.

Then you make your call.

[M]

It's not the right way to view the situation.



Your no 1 risk to your portfolio longevity is inflation. The best hedge against inflation is stocks. The assets that can be killed via inflation are cash related assets like bonds and cash.

So a portfolio that is 100% cash related assets has a significant chance of failure and poor real returns over the longer term.

Cash/bonds -> your short term hedge against asset deflation (stocks/property etc).
Stocks -> your long term hedge against inflation.

If you want to have a 100% cash related asset type portfolio you'd better be prepared for the cost of that portfolio which will mean getting a very low WR which means additional years working.

I have a fantastic suggestion for when people have these ideas to discuss. It should be a process:-

1. State your premise
2. Back up your premise with empirical evidence.
3. State the pros and cons of this approach.

=> So in this case the premise is stocks and bonds are overvalued and therefore we are at a tail end of the curve and therefore you require 100% cash related assets.

The evidence for being at that part of the curve is not currently available. You won't have that evidence until we get to the future point which is your focus. It could be 20 years but it could be 50 years or it could be 5 years.

The pros of this approach are that your portfolio is relatively safe over shorter time frames but relatively unsafe over longer time periods.

Then you make your call.

Oh - right.

I bonds and TIPS are inflation protected. The returns are very different from cash, especially in today's environment. I am assuming one holds these directly and to full term. Stagflationary environments are precisely what these investments are designed to protect against.

As for stock valuations, I am basing this on historical correlations between Shiller PE ratio and 10 year stock returns. See here for further reading: madfientist.com/safe-withdrawal-rate/

I am also assuming a 25 year time horizon until social security, which is my situation.

Note that I have not done this, I am actually invested in about 98% equities and have been since 2008. Considering the potential risks in the world I have considered changing strategies somewhat and putting more money into I bonds or tips, something like a 90% equities/ 10% IBonds/tips. As I have gotten closer to actually retiring my risk tolerance, especially in today's world, has gone down somewhat.

[steveo73]

I bonds and TIPS are inflation protected. The returns are very different from cash, especially in today's environment. I am assuming one holds these directly and to full term. Stagflationary environments are precisely what these investments are designed to protect against.

Assuming you want this money to get you to point x it sounds good.

As for stock valuations, I am basing this on historical correlations between Shiller PE ratio and 10 year stock returns. See here for further reading: madfientist.com/safe-withdrawal-rate/

I understand this point but there is no guarantee. There are also so many other variables in the context of early retirement I wouldn't pay much if any attention to stock valuations. I'd consider my portfolio construction though in light of this data.

Note that I have not done this, I am actually invested in about 98% equities and have been since 2008. Considering the potential risks in the world I have considered changing strategies somewhat and putting more money into I bonds or tips, something like a 90% equities/ 10% IBonds/tips. As I have gotten closer to actually retiring my risk tolerance, especially in today's world, has gone down somewhat.

This is exactly the sort of stuff to consider. I think your lifecycle phase is a lot more relevant to your portfolio construction compared to stock valuations for instance. Everyone though is entitled to their own assessment of the situation and can construct a portfolio that suits them.

I think a good idea is to point out the pros and cons of an approach. I personally wouldn't take the approach of just holding off until social security payments at a later point in time. One of my personal back-ups is that social security could fund our lifestyle with no problems at all right now. We can get social security in 20 years time. I don't want to though rely on social security. I don't expect to require it.

It's your call though. I think you are probably like me though and not comfortable with too much bonds and cash.

[prudentelo]

Even if that number or fornula isnt perfect returns should move with yields and rates and did so in the past.

Superbears overestimate danger because at very low WR portfolio is lkely to mean revert and returns go back up before losing all money. Shiller found best correlation CAPE10 to return at 10 year while WR2% is 50 year horizon.

But basic problem is real.

[steveo73]

But basic problem is real.

Sure but is it as big a problem as not being able to estimate your expenses perfectly year on year. Is it more or less of a potential problem than a world war ? What about health scares ? What about divorce ?

What about your overall plan ? Do you have back-up options ? Social Security ? Inheritance ?

You end up with this theoretical issue that is pretty trivial in the overall scheme of things.

When the solution is you come up with your own whiz bang customized portfolio you are taking on additional risk and your calculations don't have any data to back it up.

There are always flaws. No plan is perfect. It's pros and cons.

You can always increase your chances of not running out of money by getting to a lower WR but the cost is working longer.

You gotta take the emotion out because that allows you to view the situation accurately.

[xmj]

I take it there's few Argentinians / Turks / Zimbabweans / Bulgarians (etc) among us, who remember their currency drastically losing purchasing power.

There's a place for cash in a retirement portfolio, but as long as you're faced with negative real interest rates, that place is limited.

[conwy]

The MadFientist blog post (linked by M) is interesting.

Here again, though, I find an over-reliance on probabilistic thinking:

Disclaimer: It must be said that although this article will show that the SWR is safe, past performance cannot guarantee future outcomes. If you’re looking for certainty, you won’t find it here. Nothing in life is certain though except for death and…well that’s it.

I don't disagree - it's true, nothing in life is certain.

However does that mean we shouldn't take reasonable measures to protect against even unlikely possibilities? No!

I'm not saying we should all build ourselves underground bunkers in case of a nuclear missile strike. But if you're living in an area that has experienced flooding several times in the past century, it's probably not a bad idea to keep a small bag of emergency provisions in your house and check the weather forecast from time to time. A small, convenient action to mitigate an unlikely but potentially catastrophic risk seems reasonable.

If we know that there is a certain % of probability that our planned withdrawal rate isn't safe, and some reasonable measures can be taken to make it more safe, then it seems rational to taken those measures.

Holding some cash outside of our portfolio, assuming our portfolio is already large enough to meet our planned living expenses at our planned withdrawal rate after inflation, is one such measure. Holding cash means we can fund our lifestyle for a period of time during a protracted market downturn without being forced to sell our portfolio at a low price, which makes our chosen withdrawal safer than it would be otherwise.

Is there any guarantee that we will have enough cash to beat inflation and/or last long enough during a protracted downturn? No. But having no 100% guarantee shouldn't prevent us from trying to reduce our sequence of returns risk using cash.

Compare this to driving vs. walking. If you can reach your destination on foot, without driving, then statistically (all else equal) you are less likely to die if you walk. We know that driving (even safely and in a sober state) is one of the most dangerous activities in developed countries. So it is more rational to walk if possible, because it's a little safer, even if we're still 99% likely to survive driving.

If you can make an additional reasonable effort to improve your odds at success at something, rationally, it seems sensible to do so, even if it seems unnecessary because you already have good odds. If you can turn a 5% risk into a 2.5% risk, why not?

[frommi]

That whole problem of WR% is just that you guys know just one hammer for the problem and that is to buy indices.
Buy dividend growth stocks where the initial yield is 2.5-4% and that has growth of >5%, where the dividend is safely covered by earnings and just retire on dividends alone. Or buy rentals and retire on rents - maintenance costs. I dont know a lot of early retirees that do something else.

[conwy]

That whole problem of WR% is just that you guys know just one hammer for the problem and that is to buy indices.

We should think of financial assets as tools, trying to understand their individual characteristics, and then applying them as appropriate.

No single tool rules them all; it all depends on your situation and your goals.

Cash is a tool, just as indices are a tool.

If one of my goals is to fund my (very frugal) living costs, over a fixed period of time, in the most reliable manner possible, then cash seems like a pretty good tool to achieve that goal.

If I want to have exposure to long-term market up-side to fund a frugal lifestyle over a longer and indefinite period of time in the future, an index fund seems like a good tool to achieve that goal.