Calculators :How Can They Account for Dividends?

[urn2bfree]

Forgive me if I am missing some simple answer to this, but it makes my head hurt to even try to figure it out.

Dividends are wonderful. Dividends are the rocket fuel that propel our portfolios. They have favorable tax status- and depending on how big they are and where they are (Roth), dividends can be completely and totally tax free.
That makes dividends a wonderful part of our lives, but doesn't that make them a fly in the ointment of any calculations? My understanding is that with an S&P 500 index, 89% of long term gains are from reinvestment of dividends. How can the calculators account for this?

We are not doing any withdrawing yet. Our accounts have the majority of dividends paid and re-invested in taxable accounts. We have no Roth accounts. With my wife still working and me having a sizable traditional IRA, we would take a big haircut on any money I tried to put in a Roth right now. So our account is growing as one would expect-firing on all cylinders, if you will.

But what happens when we truly FIRE-- and we take away that dividend rocket fuel?

Once my wife stops working we will be drawing down and having to use portfolio funds to pay taxes and our other living expenses. With no external source of $$ in the DeAccumulation phase, the dividends are not being reinvested. But without dividends to be intermittently pumped into the portfolio, we will miss out on 89% of the historical gains. Retirement calculators like FIRECalc use historical gains. I guess Monte Carlo situations can be set at lower gain levels, but are they set 89% lower? Besides sequence of returns risk, isn't loss of the dividend bounce a threat to any portfolio's survival?



Sent from my iPad using Early Retirement Forum

 

[Meadbh]

Well, first of all, by definition, survival during withdrawal is built into historical portfolio outcomes that you see in FireCalc, and second of all, if you use the total return approach, you will be selling some securities along the way, so there will be a mixture of dividends, capital gains and income in your withdrawals.

 

[Midpack]

Taxes are very significant. But FIRECALC (and most free calculators) does not factor in taxes, you have to do that yourself. It would probably be misleading as we can't know tax rates may change in the next 20-40 years.

And I've seen the stat showing '89% of S&P 500 returns are from reinvesting dividends' over specific periods. But that includes the dividends and the compounded returns from reinvesting those dividends. The dividends themselves are the far larger part of that 89% (take a look at the chart below from the link pdf). If you spend dividends as they come in (assuming that's what you're saying), you will still get all future dividends on equity shares you still own, along with any capital appreciation. All you've given up is the compound returns on dividends not reinvested, not the dividends which will provide a nice income.

Dividends are certainly significant and you're not losing them. The compound returns on reinvesting dividends alone are not "rocket fuel."

As you spend from your portfolio, it has to come from somewhere, there's nothing magic about the (pretax) dividends...

http://individual.troweprice.com/staticFiles/Retail/Shared/PDFs/Insights/PowerOfCompounding.pdf

 

[pb4uski]

My AA is no different today that is was in the last few years I worked and I expect to keep the same 60/40 split for a long time so my dividends will be no different other than the the extent that my overall portfolio declines, but that is unlikely since the average earning rate exceeds my withdrawal rate. The more likely scenario is that our equity allocation and dividends will increase over time in total. Since we are in the 15% bracket, our dividends are not taxed.

What will happen is as we reduce our taxable account money to support our ER, the equities component of our retirement investments will gradually shift from taxable account to Roth account, but in both cases tax-free.

 

[urn2bfree]

It seems like the recovery of portfolios from drops is sped up when you add money on the dips via dividend reinvestment. So without that boost, with dividends being withdrawn, it would take longer for portfolio recovery. Since most of the historical returns seem to include reinvestment of dividends I wonder how accurate the historical recreations are for withdrawal scenarios. I am not arguing, I honestly do not understand whether this makes a difference in how much credence to place in historical calculators...


Sent from my iPad using Early Retirement Forum

 

[pb4uski]

Money is fungible... the model would compute the net cash flow for the year (expenses - cash income, like dividends) and then apply the return for that year to get the value at the end of the year. If you're taking dividends in cash then it is just less that is taken out for expenses, but the net cash flow for the year doesn't change.

 

[audreyh1]

It seems like the recovery of portfolios from drops is sped up when you add money on the dips via dividend reinvestment. So without that boost, with dividends being withdrawn, it would take longer for portfolio recovery. Since most of the historical returns seem to include reinvestment of dividends I wonder how accurate the historical recreations are for withdrawal scenarios. I am not arguing, I honestly do not understand whether this makes a difference in how much credence to place in historical calculators...


Sent from my iPad using Early Retirement Forum

Realize that the calculators use annual rebalancing. So when there is a drop in stocks, they "buy on the dip" by selling some bonds to buy stocks. And the reverse when stocks recover. So the models "buy the dips" too. You don't need dividends to buy the dips.

The historical recreations are totally accurate.

 

[jerome len]

I'm not sure dividends make up most of the gain. I'm heavy in VG tax managed balanced fund which pays a little over 2% dividends/interest annually and has gained around 7% average over the past 10 years. My taxes from this fund are low, I can take all dividents, pay taxes and still gain about 4% a year. Now, I'm not exact in this calculation but I'm close. Also, I've switched my IRA to a Roth....income from that is non taxable as well. My biggest tax problem? Social Security.....that's what happens when you build a nice nest egg for retirement. Good Luck.

 

[photoguy]

If you are looking at dividends from the S&P 500, I wouldn't worry about it because these would be included in the total return and is modeled reasonably well by calculators like FIRECALC.

However, if you took a dividend heavy approach and your portfolio was substantially different from S&P 500 on a variety of characteristics (like volatility), I'd be more careful about generalizing from simulators (that don't adjust equity makeup).

 

[Which Roger]

Somebody correct my math if I'm wrong, but it seems that a withdrawal is a withdrawal, whether it's by taking the dividend or by selling shares...

As a simple example, say that I own 20 shares of stock, worth $10 per share. I have a 5% withdrawal rate, so I need to withdraw $10. Assume the stock pays a 5% dividend. My total dividend is $10. I can withdraw my $10 by taking the dividend rather than reinvesting it, or by selling one share of stock and reinvesting the dividend, which is worth exactly one share of stock. Either way, I still own 20 shares of stock and have $10 in my pocket.

 

[hnzw_rui]

Somebody correct my math if I'm wrong, but it seems that a withdrawal is a withdrawal, whether it's by taking the dividend or by selling shares...

As a simple example, say that I own 20 shares of stock, worth $10 per share. I have a 5% withdrawal rate, so I need to withdraw $10. Assume the stock pays a 5% dividend. My total dividend is $10. I can withdraw my $10 by taking the dividend rather than reinvesting it, or by selling one share of stock and reinvesting the dividend, which is worth exactly one share of stock. Either way, I still own 20 shares of stock and have $10 in my pocket.

Of course, if you're in the 25% marginal tax bracket, you get taxed twice - 15% on the qualified dividends and another 15% LTCG assuming the stock has appreciated if it's in a taxable account. Also makes keeping track of cost basis a bit more complicated.

In a tax deferred or tax free (e.g Roth) account, doesn't really matter.

 

[Midpack]

Of course, if you're in the 25% marginal tax bracket, you get taxed twice - 15% on the qualified dividends and another 15% LTCG assuming the stock has appreciated if it's in a taxable account. Also makes keeping track of cost basis a bit more complicated.

In a tax deferred or tax free (e.g Roth) account, doesn't really matter.

True enough, but the OP was asking specifically how (most free) calculators account for dividends.

I may be mistaken or picking nits, but the OP may be confusing how dividends impact withdrawals/net returns after taxes (significant, but taxes are not factored in most free calculators) with how dividends impact withdrawals/net returns before taxes (they make very little difference to free retirement calculators, a withdrawal is a withdrawal before taxes). Again, I may be mistaken...

 

[audreyh1]

There is no tax status difference between qualified dividends and realized long-term capital gains. I'm not sure what the OP is confused about. Tax wise there is no difference between selling 4% of your long-term assets or taking 2% in qualified dividends, and selling 2% of the assets.

Total return is made up of dividends plus asset appreciation (i.e. capital gains).

If the OP doesn't realize that the dividend and capital gain history of major indices are well documented and so accurate information is readily available, that may account for some confusion.

I also got the impression that the OP may not understand the role rebalancing plays in the models used by the calculators.

 

[Animorph]

This seems like a nice little calculator:

CAGR of the Stock Market: Annualized Returns of the S&P 500

It says from 1871 through 2014 the total return of the S&P 500 was 9.11% annualized with dividends reinvested and 4.36% annualized without dividends. So 4.36/9.11 = 48% of the annualized return was due to growth, not dividends. A far cry from 89%.

By withdrawing only dividends from a portfolio you are sort of using the percent of current portfolio strategy, although the percentage is determined for you. Barring some strange dividend change, your portfolio will never go to $0. However, you will have to accept a variable yearly income. Similar to taking, say, 3% of your portfolio every year. Generally all very safe, so you can't go too wrong.

The Schiller data that many of the historical calculators use includes dividend income, which usually buys new shares at the start of each year when the portfolio is rebalanced. So dividends are accounted for, though a calculator may not provide a dividend-only withdrawal strategy.

 

[hnzw_rui]

There is no tax status difference between qualified dividends and realized long-term capital gains. I'm not sure what the OP is confused about. Tax wise there is no difference between selling 4% of your long-term assets or taking 2% in qualified dividends, and selling 2% of the assets.

Wouldn't the tax issue only be irrelevant if the funds are in tax sheltered accounts (e.g. 401k, IRA, Roth, etc)?

Otherwise, I think there will likely be a difference between the following scenarios:

1. current share price lower than cost basis (might be better to sell assets, claim the loss and reinvest dividends - just make sure you don't run afoul of wash sale rules)
2. current share price same as cost basis (tax is the same whether you sell assets or take the dividends)
3. current share price higher than cost basis (probably better to just take the dividends)

Granted, if you're in the 0-15% marginal tax bracket with LTCG/qualified dividend tax of 0%, the above scenarios don't really matter.

 

[audreyh1]

Wouldn't the tax issue only be irrelevant if the funds are in tax sheltered accounts (e.g. 401k, IRA, Roth, etc)?

Otherwise, I think there will likely be a difference between the following scenarios:

• current share price lower than cost basis (might be better to sell assets, claim the loss and reinvest dividends - just make sure you don't run afoul of wash sale rules)
• current share price same as cost basis (tax is the same whether you sell assets or take the dividends)
• current share price higher than cost basis (probably better to just take the dividends)

Granted, if you're in the 0-15% marginal tax bracket with LTCG/qualified dividend tax of 0%, the above scenarios don't really matter.

Right - I forgot about the cost basis. Taking income as realized long-term capital gains is more tax efficient than receiving qualified dividends.

I don't think your scenarios are correct though. Qualified dividends are only the same tax-wise as the amount from selling an asset when your cost basis is 0. Otherwise the sold asset will realize less taxable income than the qualified dividend or even possibly let you take a loss.

For example - in your second scenario no taxable income is realized when selling the shares - you just got back your original investment. Otherwise known as return of capital.

 

[hnzw_rui]

Right - I forgot about the cost basis. Taking income as realized long-term capital gains is more tax efficient than receiving qualified dividends.

I don't think your scenarios are correct though. Qualified dividends are only the same tax-wise as the amount from selling an asset when your cost basis is 0. Otherwise the sold asset will realize less taxable income than the qualified dividend or even possibly let you take a loss.

For example - in your second scenario no taxable income is realized when selling the shares - you just got back your original investment. Otherwise known as return of capital.

I think I might be trying to say something similar here. Although I do have a question: why would realized long-term capital gains be more tax efficient than receiving qualified dividends? Not being sarcastic or anything. Just genuinely trying to understand.

To clarify the scenarios, assuming you have qualified dividends of $10 and are in the 25% marginal tax bracket.

1. Original Purchase Price/Cost Basis = $10, Current FMV = $2. You need to pay $1.50 in taxes on qualified dividends. However, you can claim a loss of $8 ($10 - $2) so your total tax will be reduced by $2 ($8 * 0.25). Total tax: -$0.50
2. Original Purchase Price/Cost Basis = $10, Current FMV = $10. No tax if you sell the share. You still need to pay $1.50 in taxes on qualified dividends. Total tax: $1.50
3. Original Purchase Price/Cost Basis = $10, Current FMV = $20. You need to pay $1.50 in taxes on qualified dividends. You also need to pay an additional $1.50 (($20 - $10) * 0.15) in taxes for realized long-term capital gains. Total tax: $3.00

At least, this is how I understand it. Please correct me if I'm wrong.

 

[audreyh1]

I think I might be trying to say something similar here. Although I do have a question: why would realized long-term capital gains be more tax efficient than receiving qualified dividends? Not being sarcastic or anything. Just genuinely trying to understand.

To clarify the scenarios, assuming you have qualified dividends of $10 and are in the 25% marginal tax bracket.

1. Original Purchase Price/Cost Basis = $10, Current FMV = $2. You need to pay $1.50 in taxes on qualified dividends. However, you can claim a loss of $8 ($10 - $2) so your total tax will be reduced by $2 ($8 * 0.25). Total tax: -$0.50
2. Original Purchase Price/Cost Basis = $10, Current FMV = $10. No tax if you sell the share. You still need to pay $1.50 in taxes on qualified dividends. Total tax: $1.50
3. Original Purchase Price/Cost Basis = $10, Current FMV = $20. You need to pay $1.50 in taxes on qualified dividends. You also need to pay an additional $1.50 (($20 - $10) * 0.15) in taxes for realized long-term capital gains. Total tax: $3.00

At least, this is how I understand it. Please correct me if I'm wrong.

OK - so you are assuming you receive $10 in qualified dividends no matter what.

In case A you withdrew $12 from your portfolio. You received $10 in qualified dividends, but you also decided to sell some stock at a loss and take another $2 out (the proceeds)?

In case B you withdrew $20 from your portfolio. You received $10 in qualified dividends, and you also sold some stock at no gain and took another $10?

In case C you withdrew $30 from your portfolio. You received $10 in qualified dividends, and you also sold some stock at $10 gain and took the proceeds of $20.

It seems to me in talking about taking income from a mix of dividends and selling stock you are usually trying to withdraw the same target amount. That's when you compare the tax scenarios. So I don't understand what you are trying to illustrate with your example.

If your income is from 100% qualified dividends, you are going to pay 15% on the entire amount.

If your income is from 100% selling a long term asset, i.e. realizing a long term gain (or loss), the amount of tax owed will never be as high as the qualified dividend case because the cost basis won't be zero. You might even realize a loss that can offset other income and lower your taxes.

 

[hnzw_rui]

OK - so you are assuming you receive $10 in qualified dividends no matter what.

It seems to me in talking about taking income from a mix of dividends and selling stock you are usually trying to withdraw the same target amount. That's when you compare the tax scenarios. So I don't understand what you are trying to illustrate with your example.

I didn't really think my math through when I was doing the various scenarios.

Okay, revised scenarios. Yes, dividend is $10 no matter what.

• Original Purchase Price/Cost Basis = $20, Current FMV = $10. Sell stock for $10, claim $10 loss, reinvest dividend. Total tax: -$1.00 ($1.50 from qualified dividends - $2.50 from loss)
• Original Purchase Price/Cost Basis = $10, Current FMV = $10. Doesn't matter what you do. Total tax: $1.50 from qualified dividends
• Original Purchase Price/Cost Basis = $5, Current FMV = $10. Sell stock for $10, realized LTCG, reinvest dividend. Total tax: $2.75 ($1.50 from qualified dividends + $1.25 from LTCG)

If your income is from 100% qualified dividends, you are going to pay 15% on the entire amount.

If your income is from 100% selling a long term asset, i.e. realizing a long term gain (or loss), the amount of tax owed will never be as high as the qualified dividend case because the cost basis won't be zero. It could even be a loss that can offset other income and lower your taxes.

Don't you have to pay taxes on qualified dividends regardless of whether you spend it or reinvest it? Again, please do correct me if I'm wrong.

I guess my thinking is if you can realize a loss, tax-wise you're better off selling the asset and using the proceeds as income and reinvesting the dividends. If you're realizing a gain, it might be better to use the dividend for income so you don't incur taxes on both LTCG and qualified dividends.

 

[NW-Bound]

The OP asks: "How retirement calculators account for dividends?"

I think most calculators that use historical data do as FIRECalc does. They use the total return of stocks, and do not distinguish between dividends or capital gain from price increase. Money is fungible.

Regarding the statement that 89% of stock return coming from dividends, that depends on what period we talk about. Up until 1980, that is true. Shiller has data to show that stocks mostly just kept up with inflation when stripped of their dividends, up until 1980. Since then, companies have not paid out as much, and prefer to reinvest or buy back their stock with the profits. Hence, the dividend is lousy, but the capital gain due to price increase has been higher.

So, if you can live on the lousy current dividend of 1.93% of the S&P, you will be very safe, and should see your stock having no problem keeping up with inflation in the long run, if not surpassing it.