Annual Patterns of Various Asset Classes and Rebalancing.

[audreyh1]

Anyone who observes the markets over long periods of time - equities, bonds, etc., can't help but notice some annual patterns that seem to repeat most years. Sure - some years don't follow the pattern, but it's amazing how many do.

When I initially set up my withdrawal calendar, I chose using the first week of the year to do my annual withdrawal and rebalance, as it was a practical choice for several reasons:

• Most of my distributions are paid out in December with I take in cash, and from that I withdraw my annual income, and rebalance what remains back to my AA.
• Rebalancing is a taxable event for me, and so I prefer to do it in the new tax year.
• My withdrawal is based on the Dec 31 value of the portfolio.
• I run my budget based on the calendar year.

But, I have also noticed over the past two decades that this early January time is usually (but not always) favorable in terms of common market patterns:

• Most years see a "Santa Claus" rally where the equity market runs up in late December and continues into early January. In many years this is an extension of a late Q4 rally.
• Most years also see an early January bump because is seems that a lot of pension? money is invested at the first of the year.
• Bonds seem to follow the opposite pattern - interest rates seem to run up late in the year and early in the next (making bonds cheaper), and then drop to lows mid year.
• Municipal bonds have their own pattern. For some reason issuance is higher very late in the year and early in the next, and this puts downward pressure on bond prices near year end.

So these annual patterns seem to favor selling stocks and buying bonds at the start of the year, and most years equities outperform bonds.

Just an observation.

I do use "out of range" limits on my AA in case there is a year with extreme moves. But I notice much of my rebalancing is accomplished by distributions paid out in December, because if it has been a strong stock market year, distributions seem to be proportionally higher.

 

[Winemaker]

I always had a problem with AA adjustments in January. If everyone does it at the same time doesn't that in effect, change valuations? I make my adjustments at the end of 3rd quarter. Also, it seems most of the market "crashes" or "corrections" occur in October, so I might be timing as well.

 

[audreyh1]

I always had a problem with AA adjustments in January. If everyone does it at the same time doesn't that in effect, change valuations? I make my adjustments at the end of 3rd quarter. Also, it seems most of the market "crashes" or "corrections" occur in October, so I might be timing as well.

I don't think there are nearly enough people rebalancing in January to have such an effect. If there were, you would see equity dips early most Januaries, OR the "Santa Claus" rally effect would disappear in anticipation of folks rebalancing in January.

But that has not happened. Just the opposite, in fact.

My conclusion: most investors do NOT rebalance - at least not on a Jan schedule.

 

[GravitySucks]

Thanks Audrey, I was going to look this up as I was thinking what Winemaker said may be the case.
Great thing about this place, someone else has probably already done the heavy lifting.

 

[Winemaker]

My observation is the "December Selloff" that always occurs and is always blamed on tax loss harvesting, and the "January effect" that is supposedly the indicator for the year. Seems overall it's re-allocation to me. BTW, the worst month by history is September, just a humble observation. I used to get paid big money to observe.

 

[Lsbcal]

Anyone who observes the markets over long periods of time - equities, bonds, etc., can't help but notice some annual patterns that seem to repeat most years. Sure - some years don't follow the pattern, but it's amazing how many do.
...

I'll play devils advocate and say I'm not a believer in investable annual patterns. Maybe more January's are up then down but that is to be expected in a mostly up annual SP500 history.

I'd like to see the data that shows this is something one might bet on.

 

[audreyh1]

My observation is the "December Selloff" that always occurs and is always blamed on tax loss harvesting, and the "January effect" that is supposedly the indicator for the year. Seems overall it's re-allocation to me. BTW, the worst month by history is September, just a humble observation. I used to get paid big money to observe.

Yes - the worst month on average is September, followed by August and June (S&P500 1950 through 2016).

I don't notice a December selloff most years - why do you say "always occurs"? Decembers - especially the last two weeks - tend to rise. In fact, historically, December is the best performing month of the year for the S&P500, followed by November and April. If there is some blip during December, it doesn't keep December from being a the strongest month!

A December selloff might occur during those years equity markets are already down, otherwise there is no reason to tax loss harvest because people generally will have gains instead!

I do sometimes notice an April selloff as people book some gains to pay their taxes. Although April is the third best performing month of the year for the S&P500, so it must usually be a short blip.

 

[audreyh1]

I'll play devils advocate and say I'm not a believer in investable annual patterns. Maybe more January's are up then down but that is to be expected in a mostly up annual SP500 history.

I'd like to see the data that shows this is something one might bet on.

It's just the odds. Some months on average do better than other months. Doesn't predict exactly what's going to happen in any future month of course. Just like we have more up years than down, but it doesn't predict whether next year will be up or down.
January Effect, Etc: Which Months are Best (and Worst) for the Stock Market? Looks at S&P500 1950 through 2016. Dec, Nov, April best months; Sept, Aug, and June worst months in order.

More pattern fun - http://www.investopedia.com/day-trading/best-time-day-week-month-trade-stocks/
This one uses S&P500 data from 1926 to 2004 and July just surpassing Dec for best month, followed by January, and the only negative month September.

Yet more - https://www.yardeni.com/pub/stmktreturns.pdf See S&P 500 INDEX: AVERAGE PERCENT CHANGE EACH MONTH 1928-2017 July, Dec, April best months followed by Jan. Sept, May, Feb worst months.

In all of these you see on average a strong seasonal period of months with gains from Oct through Jan.

 

[Doc Holiday]

October. This is one of the peculiarly dangerous months to speculate in stocks. The others are July, January, September, April, November, May, March, June, December, August, and February.

 

[gcgang]

Another "story" was tax loss selling maxed out in November, allowing for a bounce in December and January.

 

[audreyh1]

Another "story" was tax loss selling maxed out in November, allowing for a bounce in December and January.

Interesting. Even though November is on average a pretty strong month. September is the most down month on average - quite a bit more than any other.

Personally I don't see how tax loss selling is going to have any impact after several years reaching new market highs. Where are the losses to harvest?

Should come into play during down equity years. In that case, you would expect it to accelerate as the end of the tax year approaches.

 

[audreyh1]

Another date to be aware of is Oct 31 as being the date by which most mutual funds close out their fiscal year. There might be some "window dressing" repositioning before this date - maybe losers get kicked out, winners bid up a bit.

I suppose the Oct 31 date gives enough time to get capital gains distributions calculated and paid by the end of the tax year for individual investors.

 

[audreyh1]

Since the bulk of my income comes in December when most of my mutual funds pay out distributions, I find myself hoping for any late market rally to sustain through the end of the year.

Although since I don't automatically reinvest distributions it shouldn't matter, especially since I rebalance in Jan after taking out my annual withdrawal.

I do count on the end of year distributions to do most of my rebalancing for me.

 

[Lsbcal]

Sounds like a good plan.

 

[audreyh1]

Sounds like a good plan.

That's been the way things work the last few years.

Things will be rising towards year end, and I see my rebalance triggers getting close. And I remind myself "Distributions are coming! Distributions are coming!" as I hope things are still higher in December.

 

[RobbieB]

Wouldn't you be better off taking dough all through the year instead of all at once in January?

The market is on fire. I can see taking it early if you're expecting it to tank.

Are you?

 

[audreyh1]

Wouldn't you be better off taking dough all through the year instead of all at once in January?

The market is on fire. I can see taking it early if you're expecting it to tank.

Are you?

And how would that work? You recalculate your withdrawal monthly? Based on the portfolio balance each month? And rebalance monthly? Perhaps some people do that.

I calculate how much to remove based on the Dec 31 value. I also usually need to rebalance, and I don't do that more than once a year. So I rebalance after withdrawing my annual income from the cash in my portfolio. I don't care to do it more often - it's quite a bit of work already. And taking it out of distributions paid in December means fewer taxable events.

Regardless, I don't care to leave current year funds invested in long term investments - it's for short term spending. I still have the rest of the portfolio invested which is good enough to me.

 

[Lsbcal]

Wouldn't you be better off taking dough all through the year instead of all at once in January?

...

That is what I do. I define our max equity (currently 60%) and when it gets to 61%, I sell down to 60%. But this is in Roth and IRA accounts.

If there are big gains, then waiting until Jan 1 might be OK. But then plenty of others might be doing the same thing the first trading days of the new year.

 

[RobbieB]

It doesn't seem that tough.

Lets see on 4%, lets do 1% quarterly instead of 4% all at once.

 

[audreyh1]

It doesn't seem that tough.

Lets see on 4%, lets do 1% quarterly instead of 4% all at once.

Well it would cause too many taxable events for me assuming I am rebalancing quarterly, which I don't want to do. Since I already have funds handing back money in December that I am going to pay taxes on, I don't want to reinvest that a chunk of that money and then turn around and sell in March, June and September, creating yet more capital gains.

If I'm not rebalancing, then I'm pulling the money out of the cash portion of the portfolio to put into the cash in my short-term funds. And that makes no difference.

There aren't benefits to frequent rebalancing, and a year is kind of a minimum for rebalancing to help - you need enough time for your asset classes to diverge. Studies have shown that for taxable accounts 18 months may be a better time period for rebalancing. But I stick to what works for tax and budget accounting.

 

[audreyh1]

That is what I do. I define our max equity (currently 60%) and when it gets to 61%, I sell down to 60%. But this is in Roth and IRA accounts.

If there are big gains, then waiting until Jan 1 might be OK. But then plenty of others might be doing the same thing the first trading days of the new year.

Wait - I thought you were describing rebalancing - selling from equity and buying bonds/fixed income. But you aren't buying fixed income, you are taking profits out of the portfolio - withdrawing? That's a bit different.

Is your withdrawal rate dependent on equity performance then? That's a bit confusing.

 

[RobbieB]

What does re-balancing have to do with cashing in (or out as the case may be) more often as compared to less?

You made more dough so you have to pay more tax? Are you doing the subsidized health care thing too? Need to make less dough?

I don't get it.

 

[Lsbcal]

Wait - I thought you were describing rebalancing - selling from equity and buying bonds/fixed income. But you aren't buying fixed income, you are taking profits out of the portfolio - withdrawing? That's a bit different.

Is your withdrawal rate dependent on equity performance then? That's a bit confusing.

Well I generally rebalance asset classes at the same time as keeping the equities at the max allocation. When I say rebalance, I am doing it in all asset classes including bonds. Yes, sometimes the equity sale is handy for extra spending (cash) but generally it goes into intermediate or short term bonds.

I choose to keep the max equities rather then letting them run up for quite awhile in the hopes of catching a "rebalance premium". So one could set their equities at say 55% and rebalance annually, but I just set it higher and cap it. Again, no tax consequences for me as most investable assets are in Roth or IRA's.

The equity asset classes I use are: SV (or SG), MV (or MG or SV), LV (or LG or MV), SP500 or International Large, MV or International Small. Yes, I do some gentle market timing. This is my own weird investing technique and not recommended.

 

[audreyh1]

What does re-balancing have to do with cashing in (or out as the case may be) more often as compared to less?

You made more dough so you have to pay more tax? Are you doing the subsidized health care thing too? Need to make less dough?

I don't get it.

It comes down to the mechanics of the process. If my funds spit out most of the distributions once a year, then it is more tax efficient to do the withdrawal and rebalancing once a year shortly thereafter. If instead I reinvest that money and then take it back out piecemeal rebalancing as I go, I am generating more capital gains - could even be short term cap gains not taxed as favorably. It might not even correspond to actually making more dough overall as some assets might have dropped. It gets messy.

If your funds don't generate much in distributions then this is not an issue, but if they do it can result in higher taxable income even though you don't see a corresponding increase in asset value.

Say you took out 1% from your highest funds in each of March, June and September. Then in December your mutual funds pay out 4% in capital gains distributions. Oops! Now you owe taxes on that 4% plus whatever is a taxable gain from the other 3% in sales and in Dec you have to reinvest the 3% you don't need. Alternatively you could have waited until Dec and just taken the 4% from the distributions without incurring taxes from the other 3 sales.

Remember capital gains distributions are 100% taxable yet when paid do not create a gain. The mutual funds drop in value when the distribution is paid so your net change in asset value is zero.

So basically my strategy is to use the distributions to fund my withdrawal and minimize any additional sales.

And it's better to rebalance less often anyway, IMO no more than once a year unless there has been a sudden large market change.