Won the game? Where to set AA?

[NW-Bound]

Now, 17% cash, 20% bonds with duration< 4 years. and 63% equities. I would sell more equity, but I am already at the limit of taxable income that I wish to have for this year.

Most of the equities are biased toward growing income and hopefully reasonably stable overall income for the entire portfolio.

So, we are not that far apart in terms of equities. I am currently at 67% equities, 7% bonds, and 26% cash. The total income from the portfolio, excluding my wife's 401k MFs which do not declare dividends, is 2.5% in the last 12 months. For comparison, the venerable S&P500 ETF SPY yields only 1.93%.

That 2.5% is computed with the stinkin' 26% cash included, which yields only 1.5%, so it's not too bad. Still, after inflation I am getting only 1% real return. Without cap gains, I must really cut back on expenses, even after getting SS if this persists.

 

[Lsbcal]

Unless I miss your meaning, the only way to take advantage of a shift in rates and stock yields like you mention is to wait in cash, or very short duration fixed income. Otherwise, when rates/yields shift, your principle shifts in the opposite direction. This is what is usually condemned on here as market timing.

I'm not suggesting timing. What I meant by the recent comment is that rates may gradually shift towards historical levels and during that time maybe we will be lucky to achieve a zero real rate net, hence the patience comment. Then after that rate move, we will again be back at historical levels. My bonds are probably about 4 years duration now, but I might move the overall bond duration down a bit.

I think waiting in cash or short term bonds is going to be tricky. One looses the term premium and short rates can move up swiftly as the Fed moves to squelch inflation or business excesses. In the 2004-2007 rate move, intermediate term bonds beat short term bonds. I could display a graph if somebody wants to see it. Not saying this will happen exactly the same, just that it could happen.

I find this thread interesting, but I really cannot understand the assumptions. I also strongly doubt that many of us could psychologically tolerate a 50% drawdown, let alone 60%. Most of us have been reasonably well off our entire lives. How many could seamlessly drop our spending way down for a possibly long period of time?

I am pretty good at projecting myself into future imagined situations and seeing how they might feel. My report of what is posited on this thread is that it would feel bad, very, very bad.

ha

The point of this thread was to survive any extreme drops, like those seen in the 1930's. Since you have >60% equities (like I do currently), you must also be prepared to accept major drops. So I don't quite understand your concern about a 50% drop which could occur with 60% equities depending on one's spending levels. Maybe it is the spending levels that are the concern?

All I did was set a 50% or so floor and check out spending levels with various equity percentages. Some people set no portfolio minim, so they are assuming they will not panic in a true crisis -- not a crisis like 2008 but rather something much worse. With no minim, one would have to have a cast iron stomach ... or not understand their risk tolerance, or have some income sources like SS + pensions.

Of course, FIRECalc cannot model other evasive action one might take like reducing spending for a few years until a recovery appears to be taking place.

 

[haha]

The point of this thread was to survive any extreme drops, like those seen in the 1930's. Since you have >60% equities (like I do currently), you must also be prepared to accept major drops. So I don't quite understand your concern about a 50% drop which could occur with 60% equities depending on one's spending levels.

I sense that you are saying something important, but I still am not sure that I get your meaning. So are you saying that with 60% equities, an 80%+loss in your equity holdings will cost you a 50% drop in your overall portfolio, given that your fixed income holds steady?

This is pretty much REW's asteroid. I think I am more likely to be hit in a crosswalk or killed in street crime or have my building collapse on my head in a big quake than to lose 80% of my equities' values, or worse yet, 80% of my equity income. I have 0% of what back in the 20s was called water in my portfolio. Anything can fall in price, but I own companies with oil and gas, pipelines, timber, and other things that given reasonable financing cannot disappear. No black boxes, aka indexes.

Maybe you mean something else; perhaps you can clarify.

Ha

 

[Midpack]

Yet another interesting thread, with lots of practices I should follow more rigorously. However, I probably let minimizing taxes influence my moves (far) too much. I rebalance using bands (5/25) and not 'back to nominal' as I probably should. Same issue with withdrawals, I have come to believe there's a lot to be said for % of remaining portfolio methodology. Old habits are proving hard to break...

 

[Lsbcal]

I sense that you are saying something important, but I still am not sure that I get your meaning. So are you saying that with 60% equities, an 80%+loss in your equity holdings will cost you a 50% drop in your overall portfolio, given that your fixed income holds steady?

This is pretty much REW's asteroid. I think I am more likely to be hit in a crosswalk or killed in street crime or have my building collapse on my head in a big quake than to lose 80% of my equities' values, or worse yet, 80% of my equity income. I have 0% of what back in the 20s was called water in my portfolio. Anything can fall in price, but I own companies with oil and gas, pipelines, timber, and other things that given reasonable financing cannot disappear. No black boxes, aka indexes.

Maybe you mean something else; perhaps you can clarify.

Ha

I think that you are saying that your portfolio will hold up better then the simple FIRECalc portfolio choice I employed which was made up of 5 year Treasuries and Total Stock Market. The OP was only an illustration and others will have to actually run FIRECalc to get results they can use.

FIRECalc cannot really model my current portfolio choices but I assume the model is a reasonable substitute to first order. Some may feel FIRECalc modeling is not sufficient to model their portfolio. Perhaps that is what you are saying Ha?

It is true that some stuff in index funds would disintegrate in a 1930's market which is generally the sequence that has the worst minimum in the portfolio, hence worst minimum spending. I think that FIRECalc may have modeled the 1930's with bankruptcies and all, but I don't know for sure.

I suspect most of us have some other form of huge loss avoidance. Yours is apparently to hold what you analyze is high quality securities. Mine choice isn't really the topic of this post. But I do use index funds from Vanguard and they are well diversified but not guaranteed. I don't worry about taxes at all as our portfolios are totally tax exempt or tax deferred -- somewhat of an unusual case. We would probably cut spending a lot should a 2008 decline occur.

 

[pb4uski]

Interesting thread. Even though I think I've won the game, I plan to continue dancing with the girl that brought me to ER (Her name is equities). I've been investing in equities for over 30 years. They go up, they go down and bounce around, but I'm fine with that.

Besides, if I did reduce my equity exposure, where are those funds invested? Bonds are scary as is cash due to inflation risk. I'm in the process of buying some PenFed 2% 3 year CDs but I only see that as a shelter to ride out a possible interest rate rise.

 

[daylatedollarshort]

I sense that you are saying something important, but I still am not sure that I get your meaning. So are you saying that with 60% equities, an 80%+loss in your equity holdings will cost you a 50% drop in your overall portfolio, given that your fixed income holds steady?

This is pretty much REW's asteroid. I think I am more likely to be hit in a crosswalk or killed in street crime or have my building collapse on my head in a big quake than to lose 80% of my equities' values, or worse yet, 80% of my equity income. I have 0% of what back in the 20s was called water in my portfolio. Anything can fall in price, but I own companies with oil and gas, pipelines, timber, and other things that given reasonable financing cannot disappear. No black boxes, aka indexes.

Maybe you mean something else; perhaps you can clarify.

Ha

Ha - If I remember correctly from other threads, you have individual stocks? How do you think you fared during the recession compared to total market index funds? Would you do anything differently? Can you tell us more about your stock criteria if you have time to share?

I looked at what we lost in in index funds in under a week during the start of the last sell off, and decided I didn't have the stomach for that. I sold low but things went much, much lower after that and I bought TIPS when they were very cheap, so panicking and selling low actually worked out for once. But I need a new longer term AA and I think it will be dividend stocks for equity instead of total market.

 

[haha]

I think that you are saying that your portfolio will hold up better then the simple FIRECalc portfolio choice I employed which was made up of 5 year Treasuries and Total Stock Market. The OP was only an illustration and others will have to actually run FIRECalc to get results they can use.

You still have not affirmed that my example is what you are referring to- an 80%+ loss in equities across the board.Can you say yes or no, that this is or is not what you are talking about?

Obviously the earning power of one's holdings determines what sort of loss might be expected, and more important, what this means in terms of one's survival. My portfolio is not important to me in understanding what you are getting at. It is also not important to you, as you do not plan to use anything like this and I do not plan to try to make a case for individual stock investing. There are plusses and minuses, and most advisor types feel that the minuses are much greater. Just go with that. I just couldn't imagine that anyone on this forum was seriously planning for a possible 80% equity drawdown.

If PE 10 is currently ~25, an 80% loss means PE10 would be 5 (before earnings drops enough to influence the denominator. This would not be unprecedented, it was reached in the first half of the 1920s, but before we would get there we would have a full- on communist government, or a coup or who knows what. So my advice would be to worry about illness or something else.

Ha

 

[Lsbcal]

No I am not planning on an 80% drawdown. FIRECalc allows one to set a minimum in their portfolio. That is where the 50% number comes from in my example OP. That 50% is from a combo of spending choices, AA, and historical year of retirement. It can be hit over a bad sequence of several years, not necessarily one or two years.

I really don't know what more I can say. Not looking for an argument.

 

[haha]

No I am not planning on an 80% drawdown. FIRECalc allows one to set a minimum in their portfolio. That is where the 50% number comes from in my example OP. That 50% is from a combo of spending choices, AA, and historical year of retirement. It can be hit over a bad sequence of several years, not necessarily one or two years.

I really don't know what more I can say. Not looking for an argument.

Neither am I, just clarification. I never argue about these things. I don't really care what anyone else does or thinks. I was not able to understand what you were talking about. In any case, a drawdown does not mean in any one year, it just means a lower value than some pervious value. So I assume you mean that in some time period well before the terminal date of the portfolio plan, you do foresee, or at least plan for, an 80% drawdown, even though this is not exactly clearly stated in any of your posting on the topic.

I wonder how many people have thought through this possibility? With a liquidating portfolio plan, this is essentially what is being accepted. If people accepted this as a serious possibility, there might be lot of portfolio revamping!

Ha

 

[Lsbcal]

Here is another fellow that is discussing changes in AA: Bill Bernstein: Take Risk Off The Table

BTW, I'm not trying to convince people that equities are dangerous or banging the drums for reduced equities. Just trying to get a handle on levels of risk. Nobody here should think they are doing the wrong thing if they have convinced themselves their AA is right for them.

 

[haha]

Here is another fellow that is discussing changes in AA: Bill Bernstein: Take Risk Off The Table

BTW, I'm not trying to convince people that equities are dangerous or banging the drums for reduced equities. Just trying to get a handle on levels of risk. Nobody here should think they are doing the wrong thing if they have convinced themselves their AA is right for them.

You still haven't said that I do, or do not understand the math of what you are talking about. I have been as clear as I could; if you choose to say"no comment", I'll exit this discussion. I usually don't even read this type of thread, but both your statement that I have been trying to pin down, and NW's statement about SWRs and minimax made me want to understand this discussion.

 

[brewer12345]

The challenges of figuring this stuff out is a big reason that when I feel like I have convincingly won the race for good I will be dumping money into CPI-indexed SPIAs and TIPS (assuming rates there are acceptable).

 

[NW-Bound]

Retirement calculators like FIRECalc do not go down to the details of how various segments of the US stock market would fare over the period of 1871-2013 (142 years), as historical data that finely defined was simply not available. Even with today's technology, when I wanted to look back at some stocks I owned that have been merged or bought out, I had a tough time.

Hence, for historical performance testing, usually only simplistic 2-component portfolios are available, using a stock index and a long bond index. That's all we have to look back at history.

Using a 50/50 portfolio with a 3% WR, I saw that a $1M portfolio starting in 1906 would be drawn down to $422K in 1920. Similarly, a $1M portfolio starting in 1966 would be down below $426K in 1981. The 3rd bad period was starting in 1937, and reached the lowest point of $528K in 1948.

When economic conditions were as bad as the above 3 periods, reducing WR from 3% to 2% helped, but not as much as one would think. With a 2%WR, the minimum point of the above 3 periods would be $532K, $587K, and $643K. Yes, one brought up the minimum portfolio value by about 20-25%, but at a reduction of standard living to 2/3.

 

[samclem]

Lsbcal, NW-Bound,
Thanks for the OP and the continued refinement. My takeaway: If a retiree is carrying 60% equities (and that's as fine as we can cut it, there's insufficient data to model various types), takes a 2% WR (from starting value, adjusted for inflation), and does this for 30 years, then US historical returns have, in the past, had at least one case where a person like you has had his retirement portfolio reduced in (real) value by 50%. That's good to know, and is valuable information that can prompt us to recognize this possibly small but not insignificant risk and reduce it (e.g. by modifying the withdrawal method, holdings, etc) or building a backup plan.

Every week I drive to the store to buy groceries. I know there is a small but not zero chance that something bad will happen on the way: a flat tire, a minor accident, a major accident with horrible, debilitating injuries, etc. But I go anyway, like we all would. We do what we can to reduce risk (carry a spare tire, drive carefully, have insurance), but ultimately we go because we want to move on with our lives (and running out of food is bad, too).

 

[Lsbcal]

I'm not planning on a 2% WR except in extremely bad circumstances. The OP data showed one could maintain a 4%+ withdrawal if willing to take a 50% portfolio hit at some point in the future. I do not think I made that point clear in the OP. Sorry about that. Perhaps instead of dollars I should have just shown WR in percent. I reproduced the chart in the OP with WR's here:

This year we will spend around 3.4% of the 2013 starting portfolio. My whole motivation is to keep a very nice lifestyle going into the future. So like Sam, I'm planning on driving to the supermarket while taking the standard seat belt precautions -- no need for a Humvee. My FIRECalc numbers showed I could maintain a 3.9% WR at 65% equities and a 4.1% WR at 50% equities. Again, that is for a potential worst case 50% portfolio hit. The runs were for 20 years but 30 years gave the same results because the bad stuff happens early in the sequence e.g. you retire in 1929 and get hit badly in the early 1930's.

Maybe I should show the key FIRECalc settings just in case this helps to clear up any confusion. The portfolio size and SS numbers are not the ones I used to produce the above chart (or the OP chart). The WR % chart above does use our numbers.

I could have made errors and that is part of the reason for posting this -- groups are generally stronger then individuals.

Note: the 500,000 minimum allowed portfolio is (I think) implemented in FIRECalc as a fixed dollar number i.e. it may not be inflation adjusted. However, since portfolio events tend to happen in early years of the worst simulations, this may not be too much of an issue.

Minor edit: the SS numbers should use 2013 or later. My actual simulations used 2013.

 

[haha]

I wonder about the name of this thread. If an early retiree looks at this data and actually feels it in his body, he has not won any game at all. He has actually played the wrong game.The only sensible game, given reality, would be to find another way. If you are young, immediately get a government job, preferably federal or even better, congressman. If too late for that, don't retire early, or plan to live in a 4th floor walkup on a portfolio of $5mm. Or go straight to a life of welfare and petty crime. If one is already retired, do something like Brewer mentioned, buy an index linked spia and/or join AARP and get some aggressive leadership.

How many marriages are going to survive a 50% overall portfolio loss? Especially if as people say they have no expectation of a pension, and minimum expectation of SS?

How sure can one be that the big hits always come early? Given that Firecalc is just a series of historical numbers, and not an oracle, it seems likely to me that the "early out or you are ok" is very likely an artifact of the basically benign investing and business environment in 20th century America, with maybe a meaningful boost from a very handy WW2.

No wonder that during 2008 the ER.org mantra was "keep the faith", or that some of the faithful felt injured by Lord Bogle when he said, absolutely accurately, "If you cannot stand to lose anymore, sell." He didn't mean if you are losing sleep sell, he meant if your survival was still likely, but could be severely crippled by further meaningful losses, sell. Don't forget that 2008-2009 across the board equity losses were ~50%, and what is being talked about here is an across the board equity loss of 80%+. (similar to 1929-1932)

Que cosa!

 

[aim-high]

Now, 17% cash, 20% bonds with duration< 4 years. and 63% equities. I would sell more equity, but I am already at the limit of taxable income that I wish to have for this year.

Most of the equities are biased toward growing income and hopefully reasonably stable overall income for the entire portfolio.

If I lost 1/2 of my dividend income I would survive but be hurting. However, I think that is very unlikely to happen. Which is why I am income rather than total return oriented.

Ha

I was reading on setting up a liability matching portfolio. It seems that although Bernstein recommends such, he acknowledges that in the current environment it's not possible. However, in one post I was reading of his, he seemed to indicate that one could count 1/2 their dividend income toward as liability matching.

Is that what you're doing?

 

[REWahoo]

One error I see is using a past year (2012) in your SS start date. FIRECalc only accepts current or future years - it's a known bug.

 

[Lsbcal]

One error I see is using a past year (2012) in your SS start date. FIRECalc only accepts current or future years - it's a known bug.

Good catch. I actually did use 2013 but recorded the incorrect number here. I made a note on my post, thanks.

 

[haha]

I was reading on setting up a liability matching portfolio. It seems that although Bernstein recommends such, he acknowledges that in the current environment it's not possible. However, in one post I was reading of his, he seemed to indicate that one could count 1/2 their dividend income toward as liability matching.

Is that what you're doing?

I may not know the lingo well enough to interpret correctly or to be sure I understand the question. I see liability matching as a function of a pension fund or insurance company where the cash calls on the portfolio are well known, or stochastically forecastable as in the case of P&C insurance companies. An individual gets no such help from the law of large numbers

I just try to invest in solid, dividend paying and growing basic businesses at what seem to be good prices. Then I only replace them if they get way overvalued, or if I think I see a negative change. I live cheaply; so my "taxable portfolio" income is enough to pay income taxes and my living expenses. My IRA and Roth are smaller, but I leave the Roth alone and just take my RMDs in stock and add it to my taxable portfolio. I also hedge in the taxable portfolio when my judgment is that the overall market is in "possibility of abrupt fall" territory.

I do not recommend this; but it is entertaining and very familiar to me so I feel fairly solid with it. To me, it does seem better than relying on timely stock sales to meet current expenses, but the vast majority of members here and commentators elsewhere see this issue differently, and they may well have the better plan. Is amounts to a small part time business, and likely a reasonably good part time business would pay quite a bit better.

As others have pointed out, even in this thread, this method cannot get a blessing from Firecalc, due to its idiosyncratic nature and lack of systematic data sets.

Ha

 

[NW-Bound]

To avoid confusion, I feel that I should stress here that the low WR that I quoted to guarantee that the portfolio would not drop below 50% is a lot lower than Lsbcal cited. It is because my number assumes no other income sources. For people who want to retire very early in their 30s or 40s, this makes a big difference.

If one has a 60/40 portfolio, if the market drops as much as 80% in one year, meaning stocks are down to 20c on the dollar while the bond portion stays constant, his portfolio would become 60*0.20 + 40 = 52. This is very unusual, and I am afraid there would other major disruptions that make life very difficult, let alone what the stock market does.

What FIRECalc shows, however, is that the 50% draw down does not happen overnight but after a few years or a decade of slow grinding economy which did not give much opportunity to the investor in either the bond or stock market. One can check this out by running 100% stock and 100% bond portfolios, then see how both succumbed in intervals that overlapped.

There was little point in balancing between two lousy investment vehicles when that happened.

PS. We should be thankful for big market swings. It gives one a chance to get ahead by trading between stocks and bonds!

 

[Lsbcal]

Thanks for the clarification NW-Bound. Yes, my numbers were for 2 adults both taking SS at retirement and no other income sources.

In any case, there is no substitute for running your own FIRECalc simulations. Hopefully this thread just shows some considerations in how to run FIRECalc and analysis of simulation results.

Looking at really bad markets is probably easier to do in good equity markets like now then when stocks turn down.

 

[NW-Bound]

By the way, the bad two periods when neither stock nor bond did well were roughly 1906-1920, and 1960-1980.

And a period where both stock and bond did well was 1980-2000. One should not be surprised that most of us ERs benefited from this prosperous era.

 

[ERD50]

To avoid confusion, I feel that I should stress here that the low WR that I quoted to guarantee that the portfolio would not drop below 50% is a lot lower than Lsbcal cited. It is because my number assumes no other income sources. For people who want to retire very early in their 30s or 40s, this makes a big difference. ...

Now I'm confused.

The WR from the portfolio is independent of other income sources. But of course, those other income sources will set a floor for your income. Right?

So it might be less confusing to say that spending is less affected with other income, but that the portfolio WRs are what they are?

-ERD50