Comments on C. Reed's paper, "Rethinking Asset Location"

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
User avatar
grabiner
Advisory Board
Posts: 20992
Joined: Tue Feb 20, 2007 11:58 pm
Location: Columbia, MD

Comments on C. Reed's paper, "Rethinking Asset Location"

Post by grabiner » Sat Oct 26, 2013 4:59 pm

In a thread on tax-adjusted asset allocation, and another thread on how some spreadsheets make the adjustment, a paper by C. Reed, "Rethinking Asset Location," http://papers.ssrn.com/sol3/papers.cfm? ... id=2317970, is cited. Despite the title of the paper, it covers both issues of asset location (which assets should be held in taxable versus tax-deferred accounts) and tax-adjusted asset allocation (how to value asset classes held in different accounts.)

One adjustment which this paper makes and which I have long advocated is that tax-deferred (traditional 401(k) or IRA) and tax-free (Roth) accounts are equivalent after adjustment; the tax-deferred account must be adjusted based on the percentage that the government owns. If you will pay 25% taxes on your IRA withdrawals, then an IRA with a nominal balance of $4000 represents $3000 owned by you tax-free, and $1000 owned by the government, and should be invested almost the same (see below) as if it were a Roth IRA with a balance of $3000. This is an important point which is often missed; some investors prefer to put stocks in Roth IRAs because the growth is tax-free and thus the expected final value is higher, not recognizing that this higher expected value comes with a greater risk and a larger loss if the stock market loses value.

I disagree with the paper's decision not to tax-adjust taxable accounts as well; while the nature of the tax is different (a tax on gains, rather than on the full value), it is still a concern which affects the money you will have to spend.

Key points of the paper:
This paper argues against AL practices that a) phrase general rules in terms of asset-types, instead of in terms of the metrics that decide the issue, b) produce rules for AL that are shown to fail to maximize wealth and c) ignore the impact of a change in tax rates between contribution and withdrawal, and d) make the objective of AL to maximize utility, when investors assume AL decisions maximize wealth.


Point (a) is correct by definition but hard to work with in practice. The key metrics for tax efficiency are return and tax rate. For example, while fixed income is normally tax-inefficient, low-yielding short-term bonds do not lose much to taxes and therefore belong in taxable accounts. But it's best for many investors to start with general rules, particularly when the actual answers are dependent on the exact estimates made about future tax rates and returns. And the main rules of thumb are fairly clear: stocks are tax-efficient, bonds become less tax-efficient as the interest rate rises, and it usually works out that stocks are better in taxable accounts. You do need to work out the numbers yourself; for example, Treasury bonds have a low yield and are exempt from state tax, so if you pay high state tax, you may want to hold Treasury bonds in taxable even if yu hold your other bonds in tax-deferred. (At current yields, I recommend I-Bonds, and even TIPS, in taxable accounts for investors who pay high state tax.)

I disagree with point (d) and therefore with point (b). An investor's goal should be to maximize expected utility, not expected wealth; this is why most investors do not hold 100% stock. Holding stocks in a taxable account, rather than an equal dollar amount in a tax-free account, reduces the risk. If the tax-free account loses $10,000 the year before withdrawal, that is $10,000 less to spend; if the taxable account loses $10,000 the year before withdrawal, that is $8500 less to spend (assuming 15% tax on a long-term capital gain). And if the tax-free account loses $10,000 long before withdrawal, that is a loss of retirement spending equal to the future tax-free returns on $10,000; if the taxable account loses $10,000 long before withdrawal, that is a loss of retirement spending equal to the future returns on a $10,000 taxable investment, reduced by a further $1500 of capital-gains tax that will no longer be due (and an even lesser loss if part of the loss is on recently-bought shares and can be recovered immediately from the IRS by tax loss harvesting).

My disagreement is based on the assumption that asset allocation is used to set mathematical risk levels. The author writes, "It is the investors’ emotional reaction that determines their tolerance for portfolios’ greater or lesser risk." Risk should be determined by three criteria: need, ability, and willingness to take risk. Only the willingness criterion is emotional; if you are likely to panic or lose sleep when your portfolio loses 30% of its value, then you should choose an allocation which is unlikely to lose 30%, however you measure the losses. Your need to take risk is based on the after-tax value of your portfolio, since that is what you will have available for spending. And your ability to take risk is based on the after-tax cost of any potential losses, since these are the losses you will either suffer or make up. I will update the wiki article Tax-adjusted asset allocation based on the three views of risk.

Theoretically, the proper way to deal with emotional risk is to view your accounts as having already lost any expected future tax loss, but that is difficult for most investors. A decrease of $10,000 in an account balance looks the same to most investors whether it is $10,000 in a Roth, $10,000 in a taxable account which can be immediately harvested as a loss to offset capital gains for a $1500 savings or to save $2500 by offsetting ordinary income for four years, $10,000 in a taxable account which still has gains and thus will reduce taxes by $1500 upon withdrawal (plus any taxes due on future appreciation), or $10,000 in a 401(k) which will be taxed at 25% and is thus a cost of $7500. (My own asset allocation spreadsheet is designed for this adjustment. It never adds the dollar balance of all the accounts, only the adjusted value, so a $10,000 decrease on an account taxed at 25% shows up as only a $7500 loss on the bottom line.)

Viewing risk mathematically also justifies making asset location decisions based primarily on percentage tax costs. If a particular stock and bond investment lose the same percentage of their value to taxes, putting the stock in a tax-free account will give higher after-tax wealth if the stock has higher returns. But those higher returns are not certain, which is why you held the bond in the first place rather than 100% stock. Putting the stock in a tax-free account will give the same after-tax wealth if stock and bond returns are equal (and the tax costs are still the same), and a lower after-tax wealth if the stock market loses money.

And I agree with point (c); the tax rates on contributions are only relevant in your decision whether to invest in a tax-deferred or tax-free account. Once you have made the decision, tax-deferred and tax-free accounts are equivalent.

The article suggests putting lower-returning assets into a tax-deferred rather than tax-free account, to reduce the probability that higher-returning assets will be taxed in a higher bracket than expected. The article only considered risk in one direction, but the risk-return trade-off actually works both ways to make this more desirable. If you expect to retire in a 25% tax bracket and have stocks in your tax-deferred account (with its asset allocation value reduce by 25%), and the stock market does well, you might pay 28% tax on some of your stock withdrawals, so you get less than the full value. Conversely, if the stock market does badly, you might retire in the 15% bracket and get only 15% tax savings on some of your stock losses. (However, given current tax laws, the phase-in of Social Security taxation comes at about the top of the 15% tax bracket, so if the stock market does badly, you might actually save tax at 27.75% or 46.25% rather than 15%.)
David Grabiner

DetroitRed
Posts: 177
Joined: Tue May 17, 2011 4:16 pm

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by DetroitRed » Sat Oct 26, 2013 5:36 pm

grabiner wrote:My own asset allocation spreadsheet is designed for this adjustment. It never adds the dollar balance of all the accounts, only the adjusted value, so a $10,000 decrease on an account taxed at 25% shows up as only a $7500 loss on the bottom line.


I feel like there is a market for a software program (with an easy to use interface) that would do the things Reed's paper talks about - i.e. the program could show your net worth by adjusting for "final after-tax values."

The potential market is probably not as large as for Quicken, but I bet there are a lot of people who would be interested.

Calm Man
Posts: 2917
Joined: Wed Sep 19, 2012 9:35 am

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by Calm Man » Sat Oct 26, 2013 7:51 pm

I actually read the paper as carefully as I could and Mr Grabiner's thoughts. I can follow both intuitively and I think understand what they say. That said, I really don't know what to conclude as there are so many known unknowns, the elephants being future tax rates, future taxation of capital gains and dividends, whether ROTHs are back-door taxed in some way or another, etc. So I understand the proposition advance by I think Larry Swerdroe or Rick Ferri that you should do the same allocations in all accounts. But I may well be very wrong.

User avatar
nedsaid
Posts: 8823
Joined: Fri Nov 23, 2012 12:33 pm

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by nedsaid » Sat Oct 26, 2013 10:16 pm

I agree with the idea that tax deferred vehicles like Traditional IRAs and workplace savings plans are "inflated" because taxes will need to be paid upon withdrawal. But should this cause us to change our asset allocation within our portfolios?

I think this gets into the realm of overthinking things. There are just too many unknowns in the future. This stuff is complicated enough already. I agree we should take this into account but I wouldn't tie myself into pretzels over this.

A financial planner presented to me the idea of tax diversification. That is you have money in taxable, tax free, and tax deferred accounts. This would give a retiree more flexibility in dealing with the tax bite on his or her investments. So this means to have a Roth IRA in addition to your tax deferred accounts.

As far as placement of assets in particular accounts, I am in the camp that the asset allocation for all accounts should be the same. This is particularly true of a person who has the bulk of his or her financial assets in retirement accounts. Invest the retirement accounts for maximum return. Do you want your retirement paycheck to be bigger or smaller? I think it is as simple as that.

If a person had large amounts of investments in taxable accounts compared to what they have in retirement accounts, a different strategy should be considered. Make your tax deferred accounts more bond heavy and put your REITs there since both these asset classes are tax inefficient. Make your taxable accounts more stock heavy since you can do tax loss harvesting and take advantage of the lower capital gains tax rates. If you are in a high enough tax bracket, use tax-free bonds in your taxable accounts. (Of course a pretty good argument could be made that even for an investor with large taxable and tax deferred portfolios, that the portfolios should still be allocated the same. Use tax free bonds in taxable accounts and taxable bonds, REITs, and TIPS for tax deferred accounts).

But pretty much, what goes into tax deferred accounts are deferred wages. You get the tax break on the monies going in and you pay the same tax rates as wages when the money goes out. You can put stocks in tax deferred accounts because you are precluded from getting capital gains tax treatment anyway. You may as well invest for maximum return.

So take the effect of taxes on your investments into consideration. Don't tie yourself into a pretzel over this.
A fool and his money are good for business.

IlliniDave
Posts: 2126
Joined: Fri May 17, 2013 7:09 am

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by IlliniDave » Sun Oct 27, 2013 7:30 am

Calm Man wrote:I actually read the paper as carefully as I could and Mr Grabiner's thoughts. I can follow both intuitively and I think understand what they say. That said, I really don't know what to conclude as there are so many known unknowns, the elephants being future tax rates, future taxation of capital gains and dividends, whether ROTHs are back-door taxed in some way or another, etc. So I understand the proposition advance by I think Larry Swerdroe or Rick Ferri that you should do the same allocations in all accounts. But I may well be very wrong.


One thing I do is I maintain a cash-out net worth calculation that considers tax liability on withdrawing 401k assets (and penalty tax because of my age). I do it as a single withdrawal so it's conservative (higher tax bracket). It's a sobering thing to see. But I tend to agree that beyond that, one is adding a lot of complexity that carries a lot of uncertainty. Maybe it would be more clear just what the concrete benefit is if I saw a detailed example (I haven't read the paper yet). I agree on an intuitive level it makes sense. But, it's hard for me to get married to precise AAs. Intuitively 50/50 and 60/40 don't seem all that different. I wonder how much an apparent allocation would change due to this tax adjustment for a joe average investor. If it took extreme circumstances for it to exceed a 5-10% shift, I'd have a hard time worrying about it. The "utility" idea DG mentions is an interesting one I'll have to think through some more...
Don't do something. Just stand there!

less
Posts: 158
Joined: Mon Sep 16, 2013 12:38 pm

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by less » Sun Oct 27, 2013 11:42 am

grabiner wrote:
1) Despite the title of the paper, it covers both issues of asset location (AL) and asset allocation (AA).

2) Point A (against general rules in terms of asset-types, instead of in terms of the metrics that decide the issue) is correct by definition but hard to work with in practice... And the main rules of thumb are fairly clear.

3) Point C (against ignoring the impact of a change in tax rates between contribution and withdrawal) ---- The tax rates on contributions are only relevant in your decision whether to invest in a tax-deferred or tax-free account. Once you have made the decision, tax-deferred and tax-free accounts are equivalent.

4) Point D (against making the objective of AL to maximize utility, when investors assume AL decisions maximize wealth). An investor's goal should be to maximize expected utility, not expected wealth.

5) The article suggests putting lower-returning assets into a tax-deferred rather than tax-free account, to reduce the probability that higher-returning assets will be taxed in a higher bracket than expected.


(1) The title correctly identifies the topic of the paper as Asset Location (AL). The paper does not address Asset Allocation (AA), except during an example showing process on page 12. There is essentially no discussion of AA. Even if the OP thinks Asset Allocation should be done differently in the example, it would have no effect at all on any of the rest of the paper.

(2) The fact that "the main rules of thumb are fairly clear" is not a justification for their use when they are proven to be wrong. There is no excuse for replacing complexity with falsehood, just because people prefer simplicity and want to told what to do in 10 words or less.

(3) The paper provides math proofs of the different benefits from the retirement accounts. This includes the bonus/penalty from a change in tax rates for TDAs. Why does the OP decide this differential effect does not exist ... that "tax-deferred and tax-free accounts are equivalent" ???

The paper explains how different AL decisions can increase the bonus or decrease the penalty. What is accomplished by pretending an effect does not exist?

(4) The paper does not dispute the OP's idea that the objective of investing is to maximize utility, not wealth. The paper's discussion is limited to Asset Location. The paper argues that AA can/should deal with all the risk-management issues --- leaving the objective of the AL decision to maximize the benefits from those accounts.

The OP continues the strategy of William Reichenstein to 'not mention' that his AL conclusions completely ignore any measurement/maximizing of the retirement accounts' benefits. In the name of risk-management (that they claim cannot be accomplished without allowing the AA process to dictate the AL decision) they throw away the benefits of the accounts without any measurement of the opportunity cost of doing so.

Conclusion:
* The OP misread the paper through a lens of AA. The paper was about Asset Location, not AA.
* He simply ignored the math proofs - proofs that the rules of thumb are wrong, proof that the benefits from TDAs and TEAs are different, proofs that different AL decisions can maximize benefits.
* He uses his idea (that for AA purposes the taxable account should be discounted) as an excuse to dismiss all the rest of the paper (which had nothing to do with that idea).

sesq
Posts: 544
Joined: Wed Jan 30, 2013 9:24 am

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by sesq » Sun Oct 27, 2013 12:38 pm

I compute a liability in my net worth calc for future taxes on my tax def balance. Currently my investments are only in tax deferred and tax free accounts. I contribute to a roth 401(k) and backdoor IRAs, and receive a match which is tax deferred (as is a historical balance). I thought about it a bit and concluded that since the interest on bonds would be ordinary there was more of a logical match to have those in a tax def account. So I sold all the bonds in my Roth IRA's and bumped up my bond allocation in the 401(k) (my plan won't let me direct the Roth and tax deferred balances differently). I then computed a tax adjusted balance sheet (I use 30% as my blended state/fed rate/means testing rate), and my nominal allocation is currently 81/19. Tax adjusted its 83/17. Its not a huge difference. I am 41, so I expect to start ramping up my bond allocation over the next few years.

I see the risk of loss in tax free accounts as a valid concern. I did feel some angst in 2008 when my tax free dollars dropped and Uncle Sam was not there to share in my loss. I had done past conversions and sent extra money in to the gov't only to have to eat a decline. But I guess my sense of long haul expected returns being greater for equities I am sticking with putting the bonds in tax deferred.

less
Posts: 158
Joined: Mon Sep 16, 2013 12:38 pm

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by less » Sun Oct 27, 2013 4:01 pm

May I suggest a more productive structure for disagreement.
1) Do you agree that the rule-of-thumb to "prioritize income taxed at the highest % tax rate in the shelters" was dis-proven. If not, then where are the errors in the math proof?
2) Do you agree that the rule-of-thumb to "prioritize assets attracting the highest $$ taxes (rate of return times tax rate) in the shelters" was dis-proven. If not, then where are the errors in the math proof?
3) Do you agree with the deconstruction of the differing benefits from TDAs and TEAs? If not, then where are the errors in the math proof?
4) Do agree that the common benefit from sheltering profits from tax is a function of the tradeoff from the tax rate and rate of return, according to the math in 3) and shown in the figure on page 3? If not why not?
5) Do you agree that the TDA's bonus/penalty from a change in tax rates, according to the math in 3) will result in different Asset Location conclusions depending on your expectations? Figure 1 shows that benefits are higher when Asset A is in the tax shelters (when rates of cont and withdrawal are the same). But change the variables to contributions at 20% and withdrawals at 40% and you see that Asset B in the tax shelter gives the better result for TDAs - the reverse outcome from the TEA. If not what is wrong with the math proof?
6) Do you believe that the Asset Allocation process should manage your risk profile, and Asset Location should maximize the account's benefits? Or do you believe both AA and AL together are necessary to manage risk, and the account's benefits should be ignored?
Last edited by less on Thu Oct 31, 2013 2:24 pm, edited 1 time in total.

User avatar
Electron
Posts: 1670
Joined: Sat Mar 10, 2007 8:46 pm

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by Electron » Wed Oct 30, 2013 6:44 pm

I've been quite interested in the possibility that stocks might return more in a sheltered account with bonds in a taxable account. I decided to model a few cases in a spreadsheet and see what I could learn.

The first pass would have no rebalancing and start with a lump sum and no additional investments. Asset Allocation would be 50% stocks and 50% bonds. The spreadsheet would geneate a ratio such as 1.05 for Stocks Taxable and Bonds Sheltered versus the reverse case.

Assumptions: Stock Index Fund with variable dividend payout, Bond Index Fund, 15% tax rate for dividends and capital gains, 30% tax rate for ordinary income. Option for 0% tax bracket for qualified dividends only. No rebalancing. In most cases Stock and Bond returns would not differ by more than 2 percentage points. The funds would be held for 30 years and then closed out with taxes paid. The spreadsheet would model a Taxable Account and a Non-Deductible IRA for the Sheltered account. The Non-Deductible IRA would create a level playing field. The spreadsheet would maintain cost basis for reinvested dividends. Cost basis would also be maintained for the Non-Deductible IRA. Each acccount would be closed out at the end of 30 years with all taxes paid.

Here are a few results.

Code: Select all

Stock Return  Dividend  Bond Return  Stock in Taxable Advantage  Final Stock %

     3%          0%          3%           1.090                      52.5%
     4%          0%          4%           1.127                      53.1%
     5%          0%          5%           1.166                      53.5%

     6%          0%          4%           1.150                      66.2%
     6%          3%          4%           1.104                      64.8%
     6%          5%          4%           1.075                      63.8%

     4%          0%          6%           1.203 Shelter Helps Bonds  40.2%

Code: Select all

Stock Return  Dividend  Bond Return  Stock in Taxable Advantage  Final Stock %

     6%          3%          4%           1.204   0% Tax Dividends   67.7%
     6%          5%          4%           1.240   0% Tax Dividends   68.6%

     10%         0%          3%           1.181                      88.2%
     10%         2%          3%           1.127                      87.7%
     10%         4%          3%           1.077                      87.1%
     10%         8%          3%           0.985                      85.9%

Note that the advantage for stocks in a taxable account declines as the dividend payout increases. Capital gain distributions would have the same effect. I believe rebalancing out of stocks would also reduce the advantage for stocks in a taxable account. Lastly, adding investments every year appears to reduce the advantage. One interesting observation is that higher investment returns can result in additional advantage as a result of compounding over time. However, that can include bonds in a sheltered account. Compounding can overcome a higher tax rate at the end of the investment period.

Rebalancing with taxable gains beyond some threshold would likely argue for stocks in the sheltered account depending on investment returns and tax rates.

The two cases with the 0% tax on qualified dividends have an interesting result. A higher dividend for a given overall return results in less capital gain at the end of the period which is taxed at 15%. That results in additional advantage for having stocks in a taxable account.

One flaw in these types of spreadsheets is that they typically assume a constant return every year. It might be better to have a range of positive and negative returns at least for the case where rebalancing is included.
Electron

grok87
Posts: 7594
Joined: Tue Feb 27, 2007 9:00 pm

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by grok87 » Wed Oct 30, 2013 7:35 pm

grabiner wrote:In a thread on tax-adjusted asset allocation, and another thread on how some spreadsheets make the adjustment, a paper by C. Reed, "Rethinking Asset Location," http://papers.ssrn.com/sol3/papers.cfm? ... id=2317970, is cited. Despite the title of the paper, it covers both issues of asset location (which assets should be held in taxable versus tax-deferred accounts) and tax-adjusted asset allocation (how to value asset classes held in different accounts.)

One adjustment which this paper makes and which I have long advocated is that tax-deferred (traditional 401(k) or IRA) and tax-free (Roth) accounts are equivalent after adjustment; the tax-deferred account must be adjusted based on the percentage that the government owns. If you will pay 25% taxes on your IRA withdrawals, then an IRA with a nominal balance of $4000 represents $3000 owned by you tax-free, and $1000 owned by the government, and should be invested almost the same (see below) as if it were a Roth IRA with a balance of $3000. This is an important point which is often missed; some investors prefer to put stocks in Roth IRAs because the growth is tax-free and thus the expected final value is higher, not recognizing that this higher expected value comes with a greater risk and a larger loss if the stock market loses value.

I disagree with the paper's decision not to tax-adjust taxable accounts as well; while the nature of the tax is different (a tax on gains, rather than on the full value), it is still a concern which affects the money you will have to spend.

Key points of the paper:
This paper argues against AL practices that a) phrase general rules in terms of asset-types, instead of in terms of the metrics that decide the issue, b) produce rules for AL that are shown to fail to maximize wealth and c) ignore the impact of a change in tax rates between contribution and withdrawal, and d) make the objective of AL to maximize utility, when investors assume AL decisions maximize wealth.


Point (a) is correct by definition but hard to work with in practice. The key metrics for tax efficiency are return and tax rate. For example, while fixed income is normally tax-inefficient, low-yielding short-term bonds do not lose much to taxes and therefore belong in taxable accounts. But it's best for many investors to start with general rules, particularly when the actual answers are dependent on the exact estimates made about future tax rates and returns. And the main rules of thumb are fairly clear: stocks are tax-efficient, bonds become less tax-efficient as the interest rate rises, and it usually works out that stocks are better in taxable accounts. You do need to work out the numbers yourself; for example, Treasury bonds have a low yield and are exempt from state tax, so if you pay high state tax, you may want to hold Treasury bonds in taxable even if yu hold your other bonds in tax-deferred. (At current yields, I recommend I-Bonds, and even TIPS, in taxable accounts for investors who pay high state tax.)

I disagree with point (d) and therefore with point (b). An investor's goal should be to maximize expected utility, not expected wealth; this is why most investors do not hold 100% stock. Holding stocks in a taxable account, rather than an equal dollar amount in a tax-free account, reduces the risk. If the tax-free account loses $10,000 the year before withdrawal, that is $10,000 less to spend; if the taxable account loses $10,000 the year before withdrawal, that is $8500 less to spend (assuming 15% tax on a long-term capital gain). And if the tax-free account loses $10,000 long before withdrawal, that is a loss of retirement spending equal to the future tax-free returns on $10,000; if the taxable account loses $10,000 long before withdrawal, that is a loss of retirement spending equal to the future returns on a $10,000 taxable investment, reduced by a further $1500 of capital-gains tax that will no longer be due (and an even lesser loss if part of the loss is on recently-bought shares and can be recovered immediately from the IRS by tax loss harvesting).

My disagreement is based on the assumption that asset allocation is used to set mathematical risk levels. The author writes, "It is the investors’ emotional reaction that determines their tolerance for portfolios’ greater or lesser risk." Risk should be determined by three criteria: need, ability, and willingness to take risk. Only the willingness criterion is emotional; if you are likely to panic or lose sleep when your portfolio loses 30% of its value, then you should choose an allocation which is unlikely to lose 30%, however you measure the losses. Your need to take risk is based on the after-tax value of your portfolio, since that is what you will have available for spending. And your ability to take risk is based on the after-tax cost of any potential losses, since these are the losses you will either suffer or make up. I will update the wiki article Tax-adjusted asset allocation based on the three views of risk.

Theoretically, the proper way to deal with emotional risk is to view your accounts as having already lost any expected future tax loss, but that is difficult for most investors. A decrease of $10,000 in an account balance looks the same to most investors whether it is $10,000 in a Roth, $10,000 in a taxable account which can be immediately harvested as a loss to offset capital gains for a $1500 savings or to save $2500 by offsetting ordinary income for four years, $10,000 in a taxable account which still has gains and thus will reduce taxes by $1500 upon withdrawal (plus any taxes due on future appreciation), or $10,000 in a 401(k) which will be taxed at 25% and is thus a cost of $7500. (My own asset allocation spreadsheet is designed for this adjustment. It never adds the dollar balance of all the accounts, only the adjusted value, so a $10,000 decrease on an account taxed at 25% shows up as only a $7500 loss on the bottom line.)

Viewing risk mathematically also justifies making asset location decisions based primarily on percentage tax costs. If a particular stock and bond investment lose the same percentage of their value to taxes, putting the stock in a tax-free account will give higher after-tax wealth if the stock has higher returns. But those higher returns are not certain, which is why you held the bond in the first place rather than 100% stock. Putting the stock in a tax-free account will give the same after-tax wealth if stock and bond returns are equal (and the tax costs are still the same), and a lower after-tax wealth if the stock market loses money.

And I agree with point (c); the tax rates on contributions are only relevant in your decision whether to invest in a tax-deferred or tax-free account. Once you have made the decision, tax-deferred and tax-free accounts are equivalent.

The article suggests putting lower-returning assets into a tax-deferred rather than tax-free account, to reduce the probability that higher-returning assets will be taxed in a higher bracket than expected. The article only considered risk in one direction, but the risk-return trade-off actually works both ways to make this more desirable. If you expect to retire in a 25% tax bracket and have stocks in your tax-deferred account (with its asset allocation value reduce by 25%), and the stock market does well, you might pay 28% tax on some of your stock withdrawals, so you get less than the full value. Conversely, if the stock market does badly, you might retire in the 15% bracket and get only 15% tax savings on some of your stock losses. (However, given current tax laws, the phase-in of Social Security taxation comes at about the top of the 15% tax bracket, so if the stock market does badly, you might actually save tax at 27.75% or 46.25% rather than 15%.)

THanks for posting David. I will read the paper when I get a chance- hopefully this weekend.
cheers,
"...people always live for ever when there is any annuity to be paid them"- Jane Austen

Dulocracy
Posts: 2253
Joined: Wed Feb 27, 2013 1:03 pm
Location: Atlanta, GA

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by Dulocracy » Thu Oct 31, 2013 1:14 pm

less wrote:
Conclusion:
* The OP misread the paper through a lens of AA. The paper was about Asset Location, not AA.
* He simply ignored the math proofs - proofs that the rules of thumb are wrong, proof that the benefits from TDAs and TEAs are different, proofs that different AL decisions can maximize benefits.
* He uses his idea (that for AA purposes the taxable account should be discounted) as an excuse to dismiss all the rest of the paper (which had nothing to do with that idea).


less,
This board is and has been a cordial place. The user grabiner has over 10,000 posts, and he is someone who makes people think. That is not to say that this member is always right. There are those who know a great deal who are going to miss the point and be wrong from time to time. Thank you for taking the time to post thoughtful rebuttal (something that is needed for a healthy board). That being said, the tone of your conclusion sounds like 1) you took the criticism of the article as a personal attack and/or 2) you are attacking grabiner personally. "He simply ignored the math proofs..." Perhaps he understood them in a different way. Perhaps he did not understand them. Perhaps the math proofs are incorrect. Perhaps he ignored them. That being said, you are more effective at persuasive speech when the reader/listener does not think you are attacking someone.

As to the discussion as a whole, my irrational fear of the future tax situation has caused me to predominantly use ROTH space in my retirement account. I have my ROTH and non-ROTH divided evenly. Each holds a similar asset allocation. My taxable is 100% equities, but I am going to be adding a Muni-bond fund soon. As taxable is far less than tax-free/tax-deferred for me, I thought that using Muni-bonds would be a good way of diversifying bonds while still keeping the ratio simple. For me, the math is easier if I have the same asset allocation in ROTH and traditional because my wife's 401k has all of our bonds. (She has expense ratios that range from .86 to 1.90. Because her funds are so bad, we have all of our bonds in her account and my account makes up for things like her complete lack of an international fund below an expense ratio of 1.2.) I realize, however, that the money that is in the ROTH will be mine (absent changes), and the money in the Traditional will come at a reduced rate. The reason that I keep the ratio balanced between both is that figuring out the right mix adjusting for tax rate in addition to different accounts with different rates of savings just sounded like something that was adding too much complexity.

I will admit to having a different strategy in taxable than in tax-free/tax-deferred, but that is largely because I do not have enough money there to have the same number of funds at present, and muni-bonds obviously are not the same as the bonds I would have in the retirement accounts. In taxable, we do follow the same general percentages, however (bond, international, large cap US, small cap US).
I'm not a financial professional. Post is info only & not legal advice. No attorney-client relationship exists with reader. Scrutinize my ideas as if you spoke with a guy at a bar. I may be wrong.

User avatar
Ketawa
Posts: 1888
Joined: Mon Aug 22, 2011 1:11 am
Location: DC

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by Ketawa » Thu Oct 31, 2013 1:47 pm

less wrote:5) Do you agree that the TDA's bonus/penalty from a change in tax rates, according to the math in 3) will result in different Asset Location conclusions depending on your expectations? Figure 1 shows that benefits are higher when Asset A is in the tax shelters (when rates of cont and withdrawal are the same). But change the variables to contributions at 20% and withdrawals at 40% and you see that Asset B in the tax shelter gives the better result for TDAs - the reverse outcome from the TEA. If not what is wrong with the math proof?


I disagree with this. It is also at odds with some of the other things you have written. Grabiner wrote, summarizing part of your paper:

grabiner wrote:One adjustment which this paper makes and which I have long advocated is that tax-deferred (traditional 401(k) or IRA) and tax-free (Roth) accounts are equivalent after adjustment; the tax-deferred account must be adjusted based on the percentage that the government owns.


If contributions are taxed at 20% and withdrawals are taxed at 40%, then the investor should obviously use Roth accounts. Some investors may be stuck using a tax-deferred account due to a lack of options in their 401k. If this is the case, valuing the account is best done by doing what you have already written. Reduce the value of the TDA by 40% since the government owns 40% of it.

Once the investor done this, the asset location decision is not affected anymore by the withdrawal or contribution tax rates. The variables that determine which asset is better in taxable or tax-advantaged are:
1. dividend/interest rate of the assets
2. unrealized capital gains growth rate of the assets
3. tax rate on dividends/income
4. tax rate on capital gains
5. time until withdrawal

User avatar
Ketawa
Posts: 1888
Joined: Mon Aug 22, 2011 1:11 am
Location: DC

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by Ketawa » Thu Oct 31, 2013 2:08 pm

I only had time to skim the paper right now. But this caught my eye:

When the marginal tax rate on withdrawal is lower (or higher) than the rate on contribution, an additional bonus (or penalty) is created. Past Asset Location literature has ignored this bonus/penalty.


One of the very first things that many investors ask is "Should I invest in a Roth IRA/401k, or a Traditional IRA/401k?" Investing literature certainly does not ignore the bonus/penalty caused by differences in tax rates. Investors are advised to choose between a Roth or Traditional account based on many things, one of which is their expected marginal tax rates at contribution and withdrawal.

It seems like the paper takes the view that investors should view differences in tax rates as a bonus/penalty. If it's a bonus, it tilts the decision towards putting the higher returning asset in tax-deferred. If it's a penalty, it tilts the decision towards putting the higher returning asset in tax-exempt or taxable. This is the wrong way to approach the analysis and is a form of mental accounting, in my mind.

Instead, first make the decision between Roth and Traditional type accounts, taking into account the expected contribution and withdrawal marginal tax rates amongst other factors. Then, reduce the value of Traditional accounts by the withdrawal tax rate to make them equivalent to Roth accounts. The asset location decision should not be mixed up in this process.

less
Posts: 158
Joined: Mon Sep 16, 2013 12:38 pm

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by less » Thu Oct 31, 2013 3:20 pm

Ketawa's reference was to #5 on the list I provided upstream viewtopic.php?f=10&t=125392&p=1838520#p1838520

The abstract of the paper listed as a problem that current AL discussions all "ignore the impact of a change in tax rates between contribution and withdrawal".

#grabiner's analysis was "The tax rates on contributions are only relevant in your decision whether to invest in a tax-deferred or tax-free account. Once you have made the decision, tax-deferred and tax-free accounts are equivalent." - without any backup reasoning or math.

My reply was "The paper provides math proofs of the different benefits from the retirement accounts. This includes the bonus/penalty from a change in tax rates for TDAs. Why does the OP decide this differential effect does not exist ... that "tax-deferred and tax-free accounts are equivalent" ??? The paper explains how different AL decisions can increase the bonus or decrease the penalty. What is accomplished by pretending an effect does not exist?"

Three sub-issues:

1) Do you agree with the deconstruction of TDAs benefits into
a) the difference between the future value of the after-tax savings compounded at the nominal rate of return vs the after-tax rate of return, and
b) the bonus/penalty equal to the $$withdrawn multiplied by the difference in tax rates between contribution and withdrawal. E.g.
Image

The issue being talked about here is the Deconstruction B

2) If so, then do you believe that at the time of purchase everyone (in every country) has the ability to foresee correctly what the eventual withdrawal tax rate will be?

IMO perfect foresight is not possible. You may have received bad advice at the start. Use of Roths may not be available. You may change jobs to one giving you a great Defined Benefit pension. You may marry someone with far higher income. You may end up being a great saver and investor with Required Draws much larger than expected. Governments may run out of money and raise the tax rates for everyone, etc, etc,

Should it be presumed that once the situation changes you can get out of the TDA before the higher tax rates start to apply? Certainly not in Canada. It is easier in US but still a large account cannot be moved to a Roth easily at low rates. So why generalize a presumption that everyone can?

3) So we have the existence of a bonus/penalty. We have people's situation such that they face a penalty. Does this result in different Asset Location advice?

Image

Open the spreadsheet and change the Contribution and withdrawal tax rates to 20% and 40%. . Comparing the Wealth at 30 years you see that if the TEA was used Asset A would have been better prioritized. (6,289 vs 5,915). But if the TDA was used Asset B would have been better prioritized (5,349 vs 5,078).

Conclusion - YES. a higher rate on withdrawals can reverse Asset Location advice.

Remember the subject here is Asset Location, not the decision which account to put new savings in.
Last edited by less on Sun Nov 03, 2013 8:20 am, edited 3 times in total.

User avatar
dratkinson
Posts: 4006
Joined: Thu Jul 26, 2007 6:23 pm
Location: Centennial CO

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by dratkinson » Thu Oct 31, 2013 3:25 pm

Sidebar.

Dulocracy wrote:...my wife's 401k has all of our bonds. (She has expense ratios that range from .86 to 1.90. Because her funds are so bad,...


In passing, I have heard of this idea. (I'm a novice investor, have no opportunity to use it so have not researched it, and so do not know the correct mechanics to make it work.)

Idea: Annual 1035 exchange.

(1) Assuming your wife has a low-cost MMF in her 401k, deposit all of her money there. Why? Minimizes annual costs and tax consequences (in case taxes are owed on the exchange).

(2) Do an annual 1035 exchange and transfer all of your wife's 401k MMF to her Vanguard rIRA (set up one if non existent). Once the 401k MMF funds are in your wife's Vanguard rIRA, buy whatever you want to round out your family portfolio.

So for the delay of ~1/2 year, your wife funds a low cost, high quality rIRA, while during that time she is minimizing costs (fees/taxes). Also, by transferring these funds, she is taking this tax-advantaged space away from her employer's control and putting it under her control. What's not to like?

If you like, you could also use this idea for yourself to improve your selection of funds.



Senior investors will need to correct my misunderstandings of this idea.

Example: Are 1035 exchanges limited to once/year, or is this a suggestion to simplify implementation?
Example: Assume with annual exchanges, that 1 Jan would be the preferred execution date?
d.r.a, not dr.a. | I'm a novice investor, you are forewarned.

User avatar
Ketawa
Posts: 1888
Joined: Mon Aug 22, 2011 1:11 am
Location: DC

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by Ketawa » Thu Oct 31, 2013 4:04 pm

Let's back this up a little. First question, do you agree with this, or not? Is it an accurate representation of your view?

grabiner wrote:One adjustment which this paper makes and which I have long advocated is that tax-deferred (traditional 401(k) or IRA) and tax-free (Roth) accounts are equivalent after adjustment; the tax-deferred account must be adjusted based on the percentage that the government owns.


Follow up question. Are you challenging the commutative property of multiplication? It's what makes it easy to compare Traditional and Roth accounts on an equivalent basis.

Traditional, tax-deferred
(Contribution Amount) * (Growth of Investment) * (1 - Withdrawal Tax Rate) = Final Wealth

Roth, tax-exempt
(Contribution Amount) * (1 - Contribution Tax Rate) * (Growth of Investment) = Final Wealth

If taxes at withdrawal and contribution are equivalent, then the investor is mostly agnostic between Roth and Traditional. A Roth is a little better for people who max out their tax-advantaged space.

Now going back to your sub-issues.

less wrote:1) Do you agree with the deconstruction of TDAs benefits into
a) the difference between the future value of the after-tax savings compounded at the nominal rate of return vs the after-tax rate of return, and
b) the bonus/penalty equal to the $$withdrawn multiplied by the difference in tax rates between contribution and withdrawal.


No, I don't. It's far easier to just look at the tax rates and compare them.

less wrote:2) If so, then do you believe that at the time of purchase everyone (in every country) has the ability to foresee correctly what the eventual withdrawal tax rate will be?

IMO perfect foresight is not possible. You may have received bad advice at the start. Use of Roths may not be available. You may change jobs to one giving you a great Defined Benefit pension. You may marry someone with far higher income. You may end up being a great saver and investor with Required Draws much larger than expected. Governments may run out of money and raise the tax rates for everyone, etc, etc,

Should it be presumed that once the situation changes you can get out of the TDA before the higher tax rates start to apply? Certainly not in Canada. It is easier in US but still a large account cannot be moved to a Roth easily at low rates. So why generalize a presumption that everyone can?


You're arguing against a straw man here. You don't have to know your exact future withdrawal tax rate. It's probably not 0%. It's probably not 50%. It's probably somewhere around 25%. Make your best guess, then apply this guess to the balance of the Traditional account.

Your examples only look at one direction, where your future taxes are higher than expected (big pension, spouse higher income, higher returns than expected, higher tax rates). If you're worried about this, then include it in your initial best guess. It's your best guess, after all. It's also possible that you'll have poor investment returns, marry a spouse who doesn't make a lot of money, or the progressiveness of the tax code continues.

If your expectation for the future does change at some point, then change the amount by which you adjust your Traditional account. This is the appropriate response to a realization that you will face penalty, not changing which account you use for bonds.

less wrote:3) So we have the existence of a bonus/penalty. We have people's situation such that they face a penalty. Does this result in different Asset Location advice?


If people know beforehand that they face a penalty, use Roth accounts. If they're stuck using Traditional accounts, adjust the value downwards by the best guess for a withdrawal tax rate. If it turns out that the future changes, change the amount of the adjustment from that point forward.

User avatar
Ketawa
Posts: 1888
Joined: Mon Aug 22, 2011 1:11 am
Location: DC

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by Ketawa » Thu Oct 31, 2013 4:34 pm

I could have been more succinct.

less wrote:#grabiner's analysis was "The tax rates on contributions are only relevant in your decision whether to invest in a tax-deferred or tax-free account. Once you have made the decision, tax-deferred and tax-free accounts are equivalent." - without any backup reasoning or math.


Commutative property of multiplication.

Traditional, tax-deferred
(Contribution Amount) * (Growth of Investment) * (1 - Withdrawal Tax Rate) = Final Wealth

Roth, tax-exempt
(Contribution Amount) * (1 - Contribution Tax Rate) * (Growth of Investment) = Final Wealth

You constantly have to make your best guess at your withdrawal tax rate when deciding between Roth and Traditional accounts. For any existing dollars in Traditional accounts, you can also use that best guess to adjust the value downwards. That is how the accounts are equivalent.

less
Posts: 158
Joined: Mon Sep 16, 2013 12:38 pm

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by less » Fri Nov 01, 2013 10:03 am

less wrote:Do you agree with the deconstruction of TDAs benefits into
a) the difference between the future value of the after-tax savings compounded at the nominal rate of return vs the after-tax rate of return, and
b) the bonus/penalty equal to the $$withdrawn multiplied by the difference in tax rates between contribution and withdrawal.

Ketawa wrote: No, I don't. It's far easier to just look at the tax rates and compare them.

To me that sounds like you cannot/will not argue the validity of the math that calculates the account's benefits. I can only conclude that you don't see the objective of the Asset Location decision to maximize those benefits. If you did then you would want to know the benefits and how to increase them.

The paper presents an argument that relies on you accepting each step in the process. If you don't agree with the basic premise of the introduction, there is no point talking about the 4th subsequent topic.

User avatar
Ketawa
Posts: 1888
Joined: Mon Aug 22, 2011 1:11 am
Location: DC

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by Ketawa » Fri Nov 01, 2013 3:23 pm

Reply consolidated in this thread.

less wrote:But in reality no one pays their taxes from their investment accounts. No one tracks their year-to-date and yearly profits on an after-tax basis. No one experiences the Taxable account's volatility as reduced by taxes. And the government's claim to a portion of profits does NOT mean they own a portion of your principal. The math that proponents rely on is shown in the Tax Adjusted Asset Allocation spreadsheet http://members.shaw.ca/PublicAccess/Tax ... cation.xls . They look at the salmon-coloured boxes. They should be looking at the blue boxes. And what difference does all this fiddling make? Not much. About as much as changing your Asset Allocation by a few percentage points. "


Many investors use tax adjustment to measure returns on an after-tax basis. Like Grabiner wrote in this thread:

grabiner wrote:Holding stocks in a taxable account, rather than an equal dollar amount in a tax-free account, reduces the risk. If the tax-free account loses $10,000 the year before withdrawal, that is $10,000 less to spend; if the taxable account loses $10,000 the year before withdrawal, that is $8500 less to spend (assuming 15% tax on a long-term capital gain). And if the tax-free account loses $10,000 long before withdrawal, that is a loss of retirement spending equal to the future tax-free returns on $10,000; if the taxable account loses $10,000 long before withdrawal, that is a loss of retirement spending equal to the future returns on a $10,000 taxable investment, reduced by a further $1500 of capital-gains tax that will no longer be due (and an even lesser loss if part of the loss is on recently-bought shares and can be recovered immediately from the IRS by tax loss harvesting).


I am also one of those investors that adjusts the value of my accounts to make them equivalent on an after-tax basis.

less wrote:You are perfectly free to rejig the Rebalancing calculations of the spreadsheet for the AA process you like. It should be easy. You will see that it makes no difference to the resulting Asset Location conclusion. The ending portfolio values will be different, but the Asset Location priority will stay the same.


less wrote:To me that sounds like you cannot/will not argue the validity of the math that calculates the account's benefits. I can only conclude that you don't see the objective of the Asset Location decision to maximize those benefits. If you did then you would want to know the benefits and how to increase them.


Now I will demonstrate that the Asset Location decision is not changed by a bonus/penalty from differences in contribution/withdrawal tax rate. The scenario is that an investor has $1000 in pre-tax income to put in either a Roth or Traditional account. The investor faces a 20% contribution tax rate and 40% withdrawal tax rate. The investor also has $700 to invest in a taxable account. We're still using the same Asset A and Asset B.

Go to your spreadsheet in the YearByYear tab. Set the following variables to simulate an investor who uses a Roth account. This investor starts with $800 in the Roth account.

Tax Rate on Contributions: 0%
Tax Rate on Withdrawals: 0%
RRSP Balance @ Start: $800
Taxable Balance @ Start: $700

The final ending wealth numbers match the numbers from your image earlier in this thread. Clearly, the spreadsheet is calculating the final wealth correctly using these different variables.

Ending Wealth with Asset A Prioritized in RRSP (Roth Account): 6289
Ending Wealth with Asset B Prioritized in RRSP (Roth Account): 5915

Now, set the following variables to simulate an investor who decides to use a Traditional account. This investor can look at the Traditional account balance and see $1000 in the account, but knows that 40% of the balance and 40% of all future earnings belong to the government. This means that we can simplify the situation to one where tax rates are 0%, and the investor has $600 to invest in the Traditional account.

Tax Rate on Contributions: 0%
Tax Rate on Withdrawals: 0%
RRSP Balance @ Start: $600
Taxable Balance @ Start: $700

Ending Wealth with Asset A Prioritized in RRSP (Traditional Account): 5357
Ending Wealth with Asset B Prioritized in RRSP (Traditional Account): 5034

Conclusion: an investor who tax adjusts properly does not face a bonus/penalty that affects the Asset Location decision. In both simulations, the investor is better off with Asset A in tax-advantaged.

(Note that there are still errors with the simulation which lead to the incorrect conclusion that Asset A should be in tax-advantaged. One reason for this is that the taxable account is not valued properly. Further explanation in this thread: http://www.bogleheads.org/forum/viewtopic.php?f=1&t=125474)

User avatar
Electron
Posts: 1670
Joined: Sat Mar 10, 2007 8:46 pm

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by Electron » Fri Nov 01, 2013 5:00 pm

Ketawa wrote:Note that there are still errors with the simulation which lead to the incorrect conclusion that Asset A should be in tax-advantaged. One reason for this is that the taxable account is not valued properly.

I started a post earlier today on that very subject.

In reviewing the YearByYear tab in Challenge.xls, it does not appear that any capital gains are deferred at all. Is that fair? I believe many of the investors on this board would invest in low turnover index funds and defer at least some gains. In some cases, taxes might only be paid on dividends. Dividends can be used for rebalancing. I do acknowledge that some investors will have significant turnover and not realize much advantage in holding stocks in a taxable account.

The spreadsheet would benefit from enhancements in several areas.

An option should be included for the types of disciplined investors that frequent this board. The spreadsheet should allow for a dividend payout percentage that can be specified. Gains should be deferred where possible.

Rebalancing in the spreadsheet does not appear to actually trigger a tax event at all. Maybe taxing the full stock gains every year was an approximation to handle both dividends and rebalancing.

Rebalancing when required should first start with any dividends paid. Rebalancing should be minimized.

Lastly, the spreadsheet might have an option to incorporate a range of positive and negative stock returns over 30 years. That would represent a more typical stock market. Rebalancing would look quite different with purchases or sales possible in a given asset category depending on the market environment.

Spreadsheet results should change significantly if taxable accounts are intelligently managed for maximum deferral of taxes.
Electron

User avatar
rob
Posts: 2840
Joined: Mon Feb 19, 2007 6:49 pm
Location: Here

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by rob » Fri Nov 01, 2013 5:24 pm

nedsaid wrote:I agree with the idea that tax deferred vehicles like Traditional IRAs and workplace savings plans are "inflated" because taxes will need to be paid upon withdrawal. But should this cause us to change our asset allocation within our portfolios?

I think this gets into the realm of overthinking things. There are just too many unknowns in the future. This stuff is complicated enough already. I agree we should take this into account but I wouldn't tie myself into pretzels over this.

Part of me agrees it's over thinking but the programmer in me just cannot leave that balloon effect in there :D ... so I reduce the value of the pre tax funds by a rough guess for my tax rate... it will be wrong for sure because it's over simplified but it's also wrong ignoring that will occur in my view. I know this is bad on so many levels but I keep Roth @100%, taxable @90% and pretax @80% - that way my allocation is kinda after tax.
| Rob | Its a dangerous business going out your front door. - J.R.R.Tolkien

User avatar
Doc
Posts: 7695
Joined: Sat Feb 24, 2007 1:10 pm
Location: Two left turns from Larry

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by Doc » Sat Nov 02, 2013 9:51 am

A few random comments.

I find it refreshing that less is challenging Grabiner's "attitude" towards math proofs. It is my believe that this is the David Grabiner who has a PhD in mathematics from what is arguably the most prestigious and least practical institution of higher learning in the Cambridge area. (OK, I'm biased. I went to the other one.)

But to this end David misses some practical factors.
If the tax-free account loses $10,000 the year before withdrawal, that is $10,000 less to spend; if the taxable account loses $10,000 the year before withdrawal, that is $8500 less to spend (assuming 15% tax on a long-term capital gain).
No, this is true only if that loss occurs at the very end of the withdrawal period. If it occurs earlier it is merely a deferral of taxes eventually owned. The exception is offset of ordinary income which is limited to $3000k or a benefit of only $300 in the 25% bracket. ($3000 x tax rate difference).

This whole concept of tax rate changes on the ROTH/"deferred tax" tradeoff is really unimportant. It is only a factor if the asset placement decision itself changes the tax rate. In some cases where you start at the top of a bracket it might have some effect but it is unlikely that it would take you through a whole bracket so at most it might change your tax rate by ~3%. This is insignificant given the other unknowns in the calculation.

David says he discounts the taxable account but I don't know if this is for future taxes (Reichenstein) or only unrealized tax on capital gains (Reed).

The biggest thing I find troubling in the whole asset allocation discussion is the amount of effort put into the tax efficiency of equity sub classes while FI sub classes are virtually ignored. Even in "normal" times short Treasuries are very likely the most tax efficient investment available with the possible exception of roulette. For tax efficiency everyone should slice and dice that TBM fund.

There have comments about the math being difficult. This needn't be true. If the three significant precision of the engineer is adequate for the decision you don't need the five significant figure of the mathematician and the math becomes much simpler. There is no justification for promulgating "rules of thumb" that are incorrect for a large percent of investors at the alter of simplicity.
A scientist looks for THE answer to a problem, an engineer looks for AN answer and lawyers ONLY have opinions. Investing is not a science.

less
Posts: 158
Joined: Mon Sep 16, 2013 12:38 pm

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by less » Sat Nov 02, 2013 12:36 pm

Warning!!!. None of this relates to the Asset Location decision. This is purely about the mechanics of the Asset Allocation calculation as shown on page 12 of the paper - not because the paper dealt with AA but obviously because it was necessary to explain what happens on page 13.

@grabiner (This is a re-post from the "Bond Fund Allocation" thread that seems to be repeated here. It responds to Ketawa's quote above from Grabiner.)
You are making decisions based on theoretical and logical arguments. Risk is about your gut.
Real life plays out differently - as you admit about yourself. (He admitted he did not pay his taxes from the Taxable investment account - which is necessary for his argument.)
Disparaging what everyone does as 'mental accounting' is just you imposing your personal opinion as a judgement over the rest of us. It is not 'an argument to show you are correct'.
Your ideas that 'risk' should be based on the theoretical 'dispersion width of final wealth' is another personal POV. Interesting, but I don't share it and it has no support in the literature.
You made no comment about the math basis for your argument that my spreadsheet shows false. http://members.shaw.ca/PublicAccess/Tax ... cation.xls .

Ketawa wrote: I am also one of those investors that adjusts the value of my accounts to make them equivalent on an after-tax basis.
There should be two distinctions.
1) The issue of reducing the amount of a Taxable account in the calculation of Asset Allocations. I believe this is what Ketawa is referring to
2) The argument giving rise to the advice to adjust the AA calculation, that argues that taxes should be paid from the investment accounts, and that taxes so paid effectively reduce the experience of portfolio volatility. This is what Grabiner was talking about. If you disprove the supporting argument (2) then you get rid of the need for (1).

@Rob
Just to clarify, beyond the "it just doesn't make much difference and is a waste of time" argument (which I would support) that applies to both Taxable and Tax Deferred accounts, I believe there is wide agreement that TDAs should be discounted in the Asset Allocation calculation. It is discounting the Taxable account that we are disputing.

I personally am fully prepared to accept that this is a personal POV issue. Each to his own. Ideas about risk are very personal. I will not support Grabiner's idea that everyone must do their Asset Allocation calculations with the discounted value of all Taxable Accounts. And if they don't then everything else they do regarding Asset Location is wrong.

Warning!!!. None of this relates to the Asset Location decision. This is purely about the mechanics of Asset Allocation calculations.

User avatar
Ketawa
Posts: 1888
Joined: Mon Aug 22, 2011 1:11 am
Location: DC

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by Ketawa » Sat Nov 02, 2013 1:00 pm

Money is fungible, so it doesn't matter where you think the money to pay taxes comes from. Nobody who is arguing for tax-adjusting a taxable account or a Traditional account is saying that the money to pay taxes has to come from that account.

No comment on this?

Ketawa wrote:Now I will demonstrate that the Asset Location decision is not changed by a bonus/penalty from differences in contribution/withdrawal tax rate. The scenario is that an investor has $1000 in pre-tax income to put in either a Roth or Traditional account. The investor faces a 20% contribution tax rate and 40% withdrawal tax rate. The investor also has $700 to invest in a taxable account. We're still using the same Asset A and Asset B.

Go to your spreadsheet in the YearByYear tab. Set the following variables to simulate an investor who uses a Roth account. This investor starts with $800 in the Roth account.

Tax Rate on Contributions: 0%
Tax Rate on Withdrawals: 0%
RRSP Balance @ Start: $800
Taxable Balance @ Start: $700

The final ending wealth numbers match the numbers from your image earlier in this thread. Clearly, the spreadsheet is calculating the final wealth correctly using these different variables.

Ending Wealth with Asset A Prioritized in RRSP (Roth Account): 6289
Ending Wealth with Asset B Prioritized in RRSP (Roth Account): 5915

Now, set the following variables to simulate an investor who decides to use a Traditional account. This investor can look at the Traditional account balance and see $1000 in the account, but knows that 40% of the balance and 40% of all future earnings belong to the government. This means that we can simplify the situation to one where tax rates are 0%, and the investor has $600 to invest in the Traditional account.

Tax Rate on Contributions: 0%
Tax Rate on Withdrawals: 0%
RRSP Balance @ Start: $600
Taxable Balance @ Start: $700

Ending Wealth with Asset A Prioritized in RRSP (Traditional Account): 5357
Ending Wealth with Asset B Prioritized in RRSP (Traditional Account): 5034

Conclusion: an investor who tax adjusts properly does not face a bonus/penalty that affects the Asset Location decision. In both simulations, the investor is better off with Asset A in tax-advantaged.

less
Posts: 158
Joined: Mon Sep 16, 2013 12:38 pm

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by less » Sat Nov 02, 2013 1:01 pm

Electron wrote: In the YearByYear tab in Challenge.xls, it does not appear that any capital gains are deferred at all.
This is the same misunderstanding as #ogd. This is a copy of the posts on the Bond Fund Allocation thread.

The tax rate used for the variable in this spreadsheet and for most every financial decision, is an effective tax rate not a statutory one. It includes federal and provincial and state taxes. The paper explicitly tells you to "Consider personal tax brackets, the marginal tax rate for each income type, and the effective tax rate when taxes are deferred until profits are realized. Weight the different tax burdens when returns come from multiple income-types. Net the dividend tax debits and credits."

The spreadsheet linked gives you the tools (tabs called Effective Capital Gains and Weighted Average Tax Rate) to generate your own personal tax rate given your own style of investing. People's difference in style resulting in different deferral periods for capital gains is another argument for NOT using rules-of-thumb.

Use the spreadsheet to to decide an American's prioritizing in retirement accounts between long-hold stocks and Treasury debt.
* assume tax rates on contribution/draw are the same 30% to make things simple.
* assume (say) 7% capital gains and 2% dividends for a 9% total return.
* Use tab "Effective Cap Gains Rate" to generate effective 7.1% tax rate from assumptions = 7% return, 30 years, 15% tax.
* Use tab "Weighted avg Tax Rate" to calc asset's 8.9% tax rate. (div taxed at 15%)
* Use tab "Year by Year" to compare assets:
Asset A = 9% return, 8.9% tax rate
Asset B = typical Treasury 2.5% return and 30% tax rate.

which Asset does better in the tax shelter? A = the equity with an ending value of $7,017 vs the debt at $6,477

And the assumption of holding an asset for 30 years without rebalancing or selling is ludicrous. Changing the holding period to 15 years results in the asset being taxed at 11.3%.
Add in the assumption most everyone seems to think - that withdrawal tax rate will be lower -say to 20%.
That results in a 21% difference between prioritizing equity ($7,851) vs debt ($6,496). In fact using those other assumptions you can reduce the tax rate on equity to Zero% and STILL equity is better in the shelter.

Electron wrote: Rebalancing in the spreadsheet does not appear to actually trigger a tax event at all.
Every third line calculates taxes. Taxes are calculate on profits. There is no tax event from rebalancing when all your taxes are already paid up. And No, the yearly paying of taxes on unrealized capital gains is not another 'error'. The difference between the statutory rate and the effective tax rate you input accounts for the difference.

Electron wrote: a) The spreadsheet should allow for a dividend payout percentage that can be specified. b) Rebalancing when required should first start with any dividends paid. c) The spreadsheet might have an option to incorporate a range of positive and negative stock returns over 30 years.
None of this will make a difference to the resulting AL choice. Has anyone ever said " Put debt in the shelters and equity outside, except if you receive dividends"? or "except if you rebalance with dividends first"? Or "except if your returns are variable"?

Dividend income is include in the Rate of Return variable. The tax rate for dividends is included in the effective tax rate variable. Rebalancing with dividends does not change the $$ rebalanced. There is no rebalancing bonus. On another thread someone said the spreadsheet was wrong because it did not rebalance based on %%, (instead of yearly). These are all just excuses to not believe what you don't want to believe - that the existing rules are wrong.
Last edited by less on Mon Nov 11, 2013 9:46 am, edited 2 times in total.

User avatar
Doc
Posts: 7695
Joined: Sat Feb 24, 2007 1:10 pm
Location: Two left turns from Larry

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by Doc » Sat Nov 02, 2013 1:20 pm

Ketawa wrote:Now I will demonstrate that the Asset Location decision is not changed by a bonus/penalty from differences in contribution/withdrawal tax rate. The scenario is that an investor has $1000 in pre-tax income to put in either a Roth or Traditional account. The investor faces a 20% contribution tax rate and 40% withdrawal tax rate. The investor also has $700 to invest in a taxable account. We're still using the same Asset A and Asset B.


It's not that hard. What the tax rate is when you make the contribution and when it is withdrawn makes no difference in asset location unless that location decision itself is the cause of the change in tax rate. Having higher yielding assets might cause you to have a higher withdrawal rate and therefore a higher marginal tax rate than a lower returning asset. But it is not very likely that such a decision will affect the rate enough to make a difference. Tax brackets re only a few percentage points apart and it is unlikely that one's RMDs are going to push you through two brackets. If they do you have enough income to play more golf and stop spending so much time on saving a few bucks in taxes.

I have to admit I stopped reading the Reed paper when he got to this part because it is just a nuance without meaning so I don't know if his bonus/penalty argument was addressing this point or if it is just an error.
A scientist looks for THE answer to a problem, an engineer looks for AN answer and lawyers ONLY have opinions. Investing is not a science.

User avatar
Ketawa
Posts: 1888
Joined: Mon Aug 22, 2011 1:11 am
Location: DC

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by Ketawa » Sat Nov 02, 2013 2:09 pm

My goal was to demonstrate the mathematical fact that the commutative property of multiplication makes Roth and Traditional accounts equal once you account for tax rates. Less has been insistent on these issues:

less wrote:You are perfectly free to rejig the Rebalancing calculations of the spreadsheet for the AA process you like. It should be easy. You will see that it makes no difference to the resulting Asset Location conclusion. The ending portfolio values will be different, but the Asset Location priority will stay the same.


less wrote:To me that sounds like you cannot/will not argue the validity of the math that calculates the account's benefits. I can only conclude that you don't see the objective of the Asset Location decision to maximize those benefits. If you did then you would want to know the benefits and how to increase them.


Setting the tax rates to 0% and putting $800 in a Roth account has the same result in the spreadsheet as contributing $1000 at a 20% contribution tax rate (obviously). Contributing $600 at a 0% tax rate for a Traditional account is the same as $1000 with a 40% withdrawal tax rate, unless less would like to start arguing with basic mathematical facts.

Keep raging against the conventional wisdom! I think I'm done with this thread.

less
Posts: 158
Joined: Mon Sep 16, 2013 12:38 pm

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by less » Sun Nov 03, 2013 9:24 am

This thread is now one week old. I am disappointed how very, very, very little of the paper was discussed. I went through all the posts above to find the comments relating to the the paper's points.

INTRODUCTION
A number of people prefer to not AL, but instead AA each account independently.

PHRASING OF ADVICE
One comment that false rules that can be clearly articulated are better than true advice that has complexity.
Two people most interested to dis/prove their own personal "equity sheltered/debt taxable" rule using their personal variables.

COMMONLY HEARD ASSET LOCATION RULES
Two said they used these rules and showed no interest that the paper's math shows them false.
Many tried very hard to find errors in the yr-by-yr-rebalancing spreadsheet.
a) Priority based on highest tax rate is wrong - no comments
b) Priority based on highest tax dollar (tax rate times rate of return) - no comments
.
THE BENEFITS OF TAX SHELTER ACCOUNTS
No comments on the math proof.
One refusal to even consider the math.
Two people listed the variables they believe are relevant, but not how to put them together in some metric to actually make the decision.

BENEFIT FROM SHELTERING PROFITS No comments

BONUS/PENALTY FROM DIFFERING TAX RATES BETWEEN CONTR/DRAW
No comments on the math proof.
Two statements that it does not exist - but no understandable back up for the assertion.
Two belief that the Asset Location choice will not change the bonus/penalty.

THE ASSET LOCATING PROCESS No comment

EXAMPLE OF THE PROCESS No comment

COMBINE ASSET LOCATION AND ASSET ALLOCATION PROCESS
A few comments on the Asset Allocation step (which was not discussed by the paper), but
no comment on the Asset Location.

THE ASSET LOCATION OBJECTIVE
No comment on the paper's position to use AL to maximize benefits - leaving AA to control risk.
One off-topic discussion of risk.

dbr
Posts: 24148
Joined: Sun Mar 04, 2007 9:50 am

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by dbr » Sun Nov 03, 2013 9:38 am

Less, I haven't commented because I have not yet read the paper carefully, and I am certainly not trying to follow all the comments in this thread.

However, what I will say is that I am not aware of any other paper that actually models the asset location problem in order to devise optimum strategies based on various criteria. Are you aware of any, present paper and thread notwithstanding? I don't recall if Reichenstein in his tax adjustment study addressed optimization of location using his utility criteria. Other than a simple calculation of current year tax cost, it is probably not to be expected that the answer is simple. Even that calculation can get different answers depending on investment performance and individual tax rate.

PS Duh, there's a bunch of references at the end, of course. Probably anyone who wants to comment should read all of them before doing so.

User avatar
JamesSFO
Posts: 2898
Joined: Thu Apr 26, 2012 10:16 pm

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by JamesSFO » Sun Nov 03, 2013 10:09 am

@op - thanks for the link, interesting read

@less - I think the paper is on the harder to grok/understand side without going to the website (http://www.retailinvestor.org/RRSPmodel.html) and downloading the spreadsheets and playing with them yourself. I think the author asks some provocative questions about Asset Location (AL) and the choice between TDA/TEA (tax deferred-trad 401K/IRA, tax exempt-Roth), and I also think his math is fundamentally sound.

That said, I do think he is overcomplicating things for the vast majority (I would hazard 90+%) of people who have almost all of their savings in tax sheltered (TDA/TEA) accounts and thus the question of AL isn't that interesting to them. If on the other hand you have built a substantial size of taxable, TDA, and TEA accounts then this becomes a more interesting question. At which point, given the gradient curves I do wonder if the entire thesis of the paper basically reduces to people making bets about future tax rates both at the governmental level and based on their accumulated wealth?

Calm Man
Posts: 2917
Joined: Wed Sep 19, 2012 9:35 am

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by Calm Man » Sun Nov 03, 2013 11:11 am

Less, I read the paper, twice. The second time was yesterday. I am not a dolt, at least I don't think. I could not see how one can use it to apply it to a situation. I thought it was a very poor paper. I don't even know frankly where it was published if it was and I do not know if it was peer reviewed (probably not). What take away did you get that I couldn't even gleam?

User avatar
Doc
Posts: 7695
Joined: Sat Feb 24, 2007 1:10 pm
Location: Two left turns from Larry

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by Doc » Sun Nov 03, 2013 1:05 pm

Calm Man wrote:Less, I read the paper, twice. The second time was yesterday. I am not a dolt, at least I don't think. I could not see how one can use it to apply it to a situation. I thought it was a very poor paper. I don't even know frankly where it was published if it was and I do not know if it was peer reviewed (probably not). What take away did you get that I couldn't even gleam?


I'm not sure that its intent was to show you how to apply it to a situation. It's intent was to show that many of the "rules of thumb" are misleading or down right wrong.

The last paragraph in the conclusion sums it up

Without any correct rules of thumb the Asset Location decision is time-consuming and technical. The
benefits of each asset-type must be measured and compared for each account type, given the individual
investor’s particulars. Investors should be made aware that general rules may be reversed when a
higher tax rate on withdrawal is anticipated. But the procedure need not be calculated each year.
Neither the types of securities preferred by an individual, nor their personal tax rates, will change much
from year to year.
A scientist looks for THE answer to a problem, an engineer looks for AN answer and lawyers ONLY have opinions. Investing is not a science.

User avatar
Electron
Posts: 1670
Joined: Sat Mar 10, 2007 8:46 pm

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by Electron » Sun Nov 03, 2013 1:52 pm

less wrote:Every third line calculates taxes. Taxes are calculated on profits. There is no tax event from rebalancing when all your taxes are already paid up.

Thanks for the explanation on the tax rates as described in the spreadsheet.

A spreadsheet might be more easily understood by the average investor if it operated in the same manner as an actual investment. In the case of stocks in a taxable account, I would add a column that keeps track of cost basis and also make provision for dividends. The dividend rate would be programmable and dividends would be reinvested after paying taxes. Cost basis would then be updated.

Rebalancing would also affect cost basis and taxes paid and it could get somewhat complicated. One could rebalance into or out of an asset category. In fact, rebalancing could result in a capital gain or a capital loss. I never thought about it before, but if stocks were in a decline and bonds were in an even steeper decline, one could rebalance out of stocks at a capital loss. Interesting case. However, the overall objective is to maximize the deferral of capital gains where possible.

At the end of 30 years, one would have the value of investments and cost basis. I think most investors would then like to calculate the after-tax value of the accounts and compare the total value for the case where the asset locations are reversed. One could then determine if stocks in a taxable account was the better choice or not.

In regards to my comment about not rebalancing, it is very helpful to start with a simple case to establish a reference point. A simple case is easier to comprehend for most investors. The results might represent the best case that could be approached if rebalancing winds up very minimal over time.

Please refer to my earlier post in this thread where data was presented from my spreadsheet. One example showed stocks and bonds each with a return of 4% over 30 years with no rebalancing. Stocks in a taxable account resulted in a benefit of 12.7% as compared with the reverse case. That is a significant figure. The initial 50% in stocks became 53.1% at the end of 30 years. If the dividend payout rate was 2% instead of 0%, the benefit reduces to 10.8%. I'd suggest that anyone interested in asset location read through the data presented and think about each case. There is a lot to comprehend.
Electron

User avatar
Doc
Posts: 7695
Joined: Sat Feb 24, 2007 1:10 pm
Location: Two left turns from Larry

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by Doc » Sun Nov 03, 2013 4:22 pm

Electron wrote:

Rebalancing would also affect cost basis and taxes paid and it could get somewhat complicated. One could rebalance into or out of an asset category. In fact, rebalancing could result in a capital gain or a capital loss. I never thought about it before, but if stocks were in a decline and bonds were in an even steeper decline, one could rebalance out of stocks at a capital loss. Interesting case. However, the overall objective is to maximize the deferral of capital gains where possible.

At the end of 30 years, one would have the value of investments and cost basis. I think most investors would then like to calculate the after-tax value of the accounts and compare the total value for the case where the asset locations are reversed. One could then determine if stocks in a taxable account was the better choice or not.


It's not very complicated to get a relative ranking of tax efficiencies based on one's own tax situation and performance estimates. You only have to calculate the effective annualized tax rate for LTCG which can be done by solving the coupon/zero bond equations. Then apply these rates to you assumed annual growth and compare to your alternate investment. The rebalancing problem goes away as does the reinvestment problem. (See note.) If you want to be more exact you can break down you equity into LTCG and distribution "tax costs" and add them together to get a little better number.

You are right that most investors want to know the future. If you insist on knowing what your account balances are going to be in 30 years the calculation does get very difficult. You have to worry about rebalancing and reinvestment and changes in tax rate. But you can make as intricate and exacting a spreadsheet as you want but you will never know if that return rate for the S&P 500 is going to be 8% or 6% or 10% for the next 30 years. The rest of the detail is insignificant in comparison.

All the Reed paper does is provide mathematical proofs that the well known "rules of thumb" are not universally valid especially in todays low interest rate environment.

Note: Investors tend to think of future value and therefore worry about reinvesting interest and dividends. Business types think about the present value of a project and therefore only worry about the cash flows in the future not whether or not they are going to reinvest them. The future value and present value equations give you exactly the same value for the ROI/YTM/IRR. Investors could save themselves a lot of angst if they would accept this simple mathematical principle.
A scientist looks for THE answer to a problem, an engineer looks for AN answer and lawyers ONLY have opinions. Investing is not a science.

Calm Man
Posts: 2917
Joined: Wed Sep 19, 2012 9:35 am

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by Calm Man » Sun Nov 03, 2013 5:35 pm

Doc wrote:
Calm Man wrote:Less, I read the paper, twice. The second time was yesterday. I am not a dolt, at least I don't think. I could not see how one can use it to apply it to a situation. I thought it was a very poor paper. I don't even know frankly where it was published if it was and I do not know if it was peer reviewed (probably not). What take away did you get that I couldn't even gleam?


I'm not sure that its intent was to show you how to apply it to a situation. It's intent was to show that many of the "rules of thumb" are misleading or down right wrong.

The last paragraph in the conclusion sums it up

Without any correct rules of thumb the Asset Location decision is time-consuming and technical. The
benefits of each asset-type must be measured and compared for each account type, given the individual
investor’s particulars. Investors should be made aware that general rules may be reversed when a
higher tax rate on withdrawal is anticipated. But the procedure need not be calculated each year.
Neither the types of securities preferred by an individual, nor their personal tax rates, will change much
from year to year.


Doc, this is true. So I have tax deferred and taxable accounts and may start some ROTHs soon. I have no clue from the paper what calculations to do. I certainly can't predict the tax rates in 5,10 and 20 years nor what happens to investment returns, particularly qualified dividends and capital gains. Who would have ever thought that Ronald Reagan would get rid of preferences for capital gains? Who would have thought that a new 3.8% investment tax would start this year or that investment returns would contribute towards phasing out deductions and exemptions? Not me.

User avatar
JamesSFO
Posts: 2898
Joined: Thu Apr 26, 2012 10:16 pm

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by JamesSFO » Sun Nov 03, 2013 9:49 pm

Calm Man wrote:Doc, this is true. So I have tax deferred and taxable accounts and may start some ROTHs soon. I have no clue from the paper what calculations to do. I certainly can't predict the tax rates in 5,10 and 20 years nor what happens to investment returns, particularly qualified dividends and capital gains. Who would have ever thought that Ronald Reagan would get rid of preferences for capital gains? Who would have thought that a new 3.8% investment tax would start this year or that investment returns would contribute towards phasing out deductions and exemptions? Not me.


This. To me the paper and spreadsheets are ultimately a confusing way of saying what is better is based on your beliefs about a mix of future tax rules/rates PLUS your personal tax outcome (e.g. higher/lower income at a later point of time).

User avatar
LH
Posts: 5490
Joined: Wed Mar 14, 2007 2:54 am

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by LH » Sun Nov 03, 2013 10:28 pm

one things about tax,

it seems that the boglehead retiree with 1 million to 3 million assets, pays little to no tax when retired, depending on how they have it structured, and how astute they are.

YDNAL
Posts: 13774
Joined: Tue Apr 10, 2007 4:04 pm
Location: Biscayne Bay

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by YDNAL » Mon Nov 04, 2013 6:21 am

grabiner wrote:In a thread on tax-adjusted asset allocation, and another thread on how some spreadsheets make the adjustment, a paper by C. Reed, "Rethinking Asset Location," http://papers.ssrn.com/sol3/papers.cfm? ... id=2317970, is cited.

A few thoughts in no particular order.
    1. We know our marginal tax rate today but can ONLY guess what it will be tomorrow - the further away tomorrow, the more difficult a guess. We don't know what changes to tax code await.
    2. Many (most?) folks save for retirement in work plans (401K, 403B, etc.), perhaps have traditional and Roth IRAs. Some may have insignificant (to matter) Taxable accounts. It would be interesting to know if anyone has seen statistics on this.
    3. The difference between a work plan (or traditional IRA) and a Roth IRA is deferred tax which we keep and save until withdrawal. With the latter, that tax is gone and doesn't grow/compound for us any longer.
    4. A Taxable account, unlike a Roth IRA, is taxed as accumulation moves forward (dividend & capital gains distributed), and then at withdrawal above the basis.
    5. Since interest rates are low, it seems *in vogue* to question asset placement - until interest rates are no longer low.
    6. Who is Mr. Reed anyways?
Landy | Be yourself, everyone else is already taken -- Oscar Wilde

User avatar
JamesSFO
Posts: 2898
Joined: Thu Apr 26, 2012 10:16 pm

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by JamesSFO » Mon Nov 04, 2013 8:52 am

YDNAL wrote:2. Many (most?) folks save for retirement in work plans (401K, 403B, etc.), perhaps have traditional and Roth IRAs. Some may have insignificant (to matter) Taxable accounts. It would be interesting to know if anyone has seen statistics on this.


For #2 - http://crr.bc.edu/wp-content/uploads/10 ... ble-11.pdf

The PDF gives some perspective, "Wealth Holdings of a Typical Household with Head Age 55-64, 2010"

Then burrow around on their website for more: http://crr.bc.edu/data/

User avatar
Doc
Posts: 7695
Joined: Sat Feb 24, 2007 1:10 pm
Location: Two left turns from Larry

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by Doc » Mon Nov 04, 2013 11:07 am

JamesSFO wrote: This. To me the paper and spreadsheets are ultimately a confusing way of saying what is better is based on your beliefs about a mix of future tax rules/rates PLUS your personal tax outcome (e.g. higher/lower income at a later point of time).


This is not my take on the paper. For me it is illustrating by mathematics that many of the widely published guidelines for asset placement are either in error or not widely applicable. Tax efficiency is dependent on the metrics of return and effective tax rate. The product of these two determines efficiency. Return is a pretty straight forward concept.
Effective tax rates are not a straightforward concept and vary greatly by individual. Furthermore they may be different in the future.

Two points on future tax rates:

1) Whether you use current or future tax rates depends on whether you consider asset allocation to be a means of maximizing future value or as a utility tool if I understand "utility" correctly.

2) Our estimate of future tax rates may not be precise but I have doubts that our estimates for the most part are going to effect the tax efficiency ranking of our alternatives.

The biggest factor in our portfolio's future value is the return and we don't know that for next year let alone 30 years in the future. We are not going to know perhaps within a factor of two what out before tax future value will be. The tax on that value is going to be a small part of the total no matter what the rate. All we can do is try to allocate our assets in an efficient manner. Predicting the exact amount we will "save" by that decision is a fruitless endeavor.
A scientist looks for THE answer to a problem, an engineer looks for AN answer and lawyers ONLY have opinions. Investing is not a science.

Calm Man
Posts: 2917
Joined: Wed Sep 19, 2012 9:35 am

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by Calm Man » Mon Nov 04, 2013 12:46 pm

LH wrote:one things about tax,

it seems that the boglehead retiree with 1 million to 3 million assets, pays little to no tax when retired, depending on how they have it structured, and how astute they are.


LH, I bet there are a whole bunch who wish this. At best this might work until age 70.5 after which the 401K and Ira assets come out beginning at about 4% per year and then increasing. If if there is a pension this gets messed up as well. And SS starts too. I suspect there is a sweet spot between retirement and 70.5 except for people with defined pensions.

User avatar
Electron
Posts: 1670
Joined: Sat Mar 10, 2007 8:46 pm

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by Electron » Mon Nov 04, 2013 3:00 pm

Doc wrote:All the Reed paper does is provide mathematical proofs that the well known "rules of thumb" are not universally valid especially in todays low interest rate environment.

The concept of mathematical proof is fine assuming that the algorithm in the spreadsheet is an accurate representation of future results in actual investment accounts. I've had some reservations as I have posted. In particular, it wasn't clear if the deferral of capital gains was maximized in taxable accounts.

I just compared my spreadsheet with challenge.xls and the result was somewhat unexpected. My spreadsheet as a first pass assumes no rebalancing and it uses a non-deductible IRA. It also calculates a ratio for ending wealth assuming that all accounts are closed out and taxes are paid. For the case of stock and bond returns both at 4%, I was able to produce the exact same result in both spreadsheets. In this case, stocks in a taxable account had a 1.079 advantage over the reverse asset locations.

However, to produce the 1.079 ratio, I had to assume the 4% stock return was all dividend return meaning no capital gains were deferred. I also had to take the ratio of pre-tax wealth values at the end of the period. My spreadsheet calculates the ratio based on after-tax values since that would be representative of what occurs with a typical withdrawal at the end of the period. My spreadsheet shows the ratio as 1.09 on an after-tax basis.

It might be very instructive if Less provided some actual examples to illustrate his points. The results could be provided as a ratio similar to that already described. As an example, show some cases where stocks provide a higher return in taxable accounts and show the actual ratios. Ratios could be pre-tax and after-tax. Then alter the stock and bond returns along with tax rates to demonstrate how the ratio might decline below 1.000. Also show how rebalancing might reduce the returns when stocks are placed in taxable accounts.
Electron

livesoft
Posts: 57250
Joined: Thu Mar 01, 2007 8:00 pm

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by livesoft » Mon Nov 04, 2013 3:11 pm

Calm Man wrote:Less, I read the paper, twice. The second time was yesterday. I am not a dolt, at least I don't think. I could not see how one can use it to apply it to a situation. I thought it was a very poor paper. I don't even know frankly where it was published if it was and I do not know if it was peer reviewed (probably not). What take away did you get that I couldn't even gleam?

I had the same reaction the first time. Folks are being too nice to this paper. It is garbage.

A journal editor would never have let this paper get out for review.

But folks should be pissed that they have had their time that was spent reviewing the paper stolen from them. That is time that they will never get back.
This signature message sponsored by sscritic: Learn to fish.

User avatar
Ketawa
Posts: 1888
Joined: Mon Aug 22, 2011 1:11 am
Location: DC

Re: Comments on C. Reed's paper, "Rethinking Asset Location"

Post by Ketawa » Mon Nov 04, 2013 3:33 pm

livesoft wrote:
Calm Man wrote:Less, I read the paper, twice. The second time was yesterday. I am not a dolt, at least I don't think. I could not see how one can use it to apply it to a situation. I thought it was a very poor paper. I don't even know frankly where it was published if it was and I do not know if it was peer reviewed (probably not). What take away did you get that I couldn't even gleam?

I had the same reaction the first time. Folks are being too nice to this paper. It is garbage.

A journal editor would never have let this paper get out for review.

But folks should be pissed that they have had their time that was spent reviewing the paper stolen from them. That is time that they will never get back.


Indeed, I spent a lot of time providing some detailed criticisms of the spreadsheet and gave examples that would seem to challenge some of the claimed proofs. The only response was:
"Many tried very hard to find errors in the yr-by-yr-rebalancing spreadsheet."
"Two statements that it does not exist - but no understandable back up for the assertion."
"Two belief that the Asset Location choice will not change the bonus/penalty."

I might continue to read the thread out of interest, but I don't think I'm going to bother with any more substantive conversation. Not worth the time.

Post Reply