529: Planning An Early Retirement Part 2

In Part 1 I laid out the basic question we’re trying to address:

Since the money inside a 529 plan grows tax-free, is there a break-even point, beyond which it’s more advantageous to invest in a 529 plan than in a regular taxable account even after paying penalty.

I showed that the tax advantage of 529 plans are indeed sufficient to overcome the plan costs and given time would overcome the adverse tax treatment and penalties for nonqualified withdrawals, albeit after 30 years or more in most instances. In addition, it should be understood that higher current tax bracket and lower fees will tilt things more towards 529 plans, while tax-efficient investments in taxable accounts will tilt things the other way.

Dimensional Fund Advisors (DFA)
For reasons that will soon become apparent, we enrolled in West Virginia’s Smart529 Select plan last year. We did this before the birth of our daughter to take advantage of my state’s recent tax law change to allow a deduction for contribution to an out-of-state plan. One key reason for selecting the particular plan was its use of DFA funds.

DFA funds are backed by research done by Nobel laureates in economics. I first learned of them via Paul Merriman and Index Fund Advisors (IFA). They are probably the best index fund managers, only Vanguard comes close. Altruist Financial Advisors has a great comparison chart that I think every do-it-yourself asset allocators should study. Unfortunately, there is a catch. DFA funds are only available through financial advisors who normally charge a fee for assets under management. If you are lucky enough to have a million dollar portfolio, consider Evanson Asset Management who charges a flat $2000 fee, or 0.2%, which is among the lowest that I’m aware of. A more realistic choice for me personally is IFA, which caters to account sizes of $100k and up. Their fee is 0.9% for the lowest tier.

In comparison, West Virginia’s Smart529 Select plan that gets you into DFA funds at an annual expense rate of 0.55% must be considered a bargain. I ran the same calculation as before but the 529 plan now enjoys a 0.35% advantage in fees in addition to the tax savings. The results are quite different.

Now the 15% future bracket beats the (hypothetical) 10% future LT cap gain rate in only three years! It beats the 5% rate in 10 years. Even the 25% future bracket beats the corresponding 15% rate in 13 years! The DFA funds may be more tax efficient than I supposed, but the extra advisor expense makes the 529 plan very appealing.

Qualified Educational Expenses
Even though the examples I have given so far have shown 529 plans to have the potential to be a stand along retirement savings vehicle, they are best used as intended, that is, for qualified education expenses. IRS defines these expenses as

… the tuition, fees, books, supplies, and equipment required for enrollment or attendance at an eligible educational institution (defined in the next column). They also include the reasonable costs of room and board for a designated beneficiary who is at least a half-time student.

When used for qualified educational expenses, the 529 plan is almost as good as a Roth IRA: there is an extra plan expense, off-set in part by possible state tax deductions. Many people plan on taking college courses in early retirement. As a matter of fact, numerous college towns around the country consistently rank as the best retirement destinations, so the 529 plan will come in handy then.

When the qualified education expense is reduced by a scholarship or fellowship, the earnings are taxable, but not subject to the 10% penalty. In this situation, the 529 plan acts like a non-deductible IRA. Unlike retirement plans, losses in the 529 plans are deductible, but subject to the 2% of AGI limit on Schedule A.

Rollover and Transfer of Designated Beneficiary
The 529 plan owns much of its flexibility to the freedom to rollover the plan or change the designated beneficiary to a family member of the old beneficiary. For this purpose, a “family member” is defined as

  1. Child or descendant of a child.
  2. Brother, sister, stepbrother, or stepsister.
  3. Father or mother or ancestor of either.
  4. Stepfather or stepmother.
  5. Son or daughter of a brother or sister.
  6. Brother or sister of father or mother.
  7. Son-in-law, daughter-in-law, father-in-law, mother-in-law, brother-in-law, or sister-in-law.
  8. The spouse of any individual listed above.
  9. First cousin.

This is a far-reaching list, especially since multiple rollover/re-designation of beneficiary is possible (limited to one per 12 months in some cases). Assets within a plan can be used to pay for the education of your children and the excess amount can be left to the generation after. It is certainly possible for one to pay for the education of a family member out of the 529 plan for a return payment as a way to sidestep the ordinary income tax and 10% penalty. These possibilities make the 529 plan an excellent multi-generational planning tool.

One point of note: many financial planners advocate the grandparents (of the student) be the owner of the 529 plan because the assets of the parents count against the student in many formulas used to calculate financial aid. For the same reason, it’s usually not a good idea for the student to be the owner of the 529 plan. With proper planning, gift tax for skip-generation contributions should not represent a problem.

Implications for Investment Choices within the 529 Plan
I’m currently the beneficiary of the newly-opened 529 plan. The intention is, of course, to switch over to my daughter later. I have all the money in the most aggressive (all equity) option. Chances are the money will be there for a while.

Most 529 plans feature age-based portfolios which are similar to target retirement funds that increase the fixed income portion as the target date approaches. The rational is that return expectations should be sacrificed for shielding from untimely market declines. One cannot argue with that logic if the 529 plan is established to pay for the education of a single beneficiary. However, when the 529 plan is viewed as a multi-generational planning tool, its time horizon is easily 20-30 years longer, so that the investments can be left in the most aggressive option, where the expected return is the highest, for much longer.

To compensate for the extra volatility, I will increase my contributions slightly (perhaps10-20%) above originally planned. This will likely result in an account value exceeding the educational expenses for one person down the road, but as I have demonstrated above, this is not a bad thing after all.

I originally planned for a 2-part series; however, one reader raised an interesting question of comparison to a variable annuity. I’ll try to address that in part 3 along with the asset protection feature of 529 plans.

529: Plan Early Retirement

A while ago, I wrote about 529 plans and new tax deductions in some states. Just to recap the basics about 529 plans:

  • Contributions are not deductible on federal returns; however, many states provide a deduction for contribution to in-state plans. Certain states even have deductions for contribution to out-of-state plans.
  • Although the state income tax deduction has an upper limit (e.g., for contributions up to $12,000), there is no overall contribution limit that I’m aware of. [You should consult a tax advisor here. This IRS publication states that the amount of contribution cannot be more than the qualified education expenses of the beneficiary, but I’m not sure how it is enforced. Most plans have an account limit of $2-300,000 per beneficiary.]
  • Inside the plan, money grows tax free. Qualified withdrawals for education purposes are not taxed.
  • Nonqualified withdrawals, however, have the earnings taxed as ordinary income with a 10% penalty levied.

This last point is what I want to address today. Two blog entries at MyPocketChange and RetireEarly started exploring this question:

Since the money inside a 529 plan grows tax-free, is there a break-even point, beyond which it’s more advantageous to invest in a 529 plan than in a regular taxable account even after paying the penalty?

Essentially, the interlocutor is asking whether the 529 plan can function as a stand alone retirement savings plan without contribution limits. This can be calculated quite easily if we make certain assumptions about the rate of return, current and future tax bracket, etc. But to gain a better understanding, let me rephrase the requirement into two separate conditions.

A. In each given year, the advantage of tax-free compounding must out-weigh the extra expenses for the 529 plan, viz.

R529 – ER529 > Rtaxable – ERtaxable – TRtaxable

Where R is the return; ER is the expense ratio; and TR is the tax expense for the taxable account. If this condition fails to hold, then compounding will only make things worse.

The current tax expense, TRtaxable, is given by,

TR = D * STportion * STrate + D * LTportion * LTrate

Where D is the distribution rate; STportion and LTportion are the fraction of D that are short term gains or long term gains and dividends; STrate and LTrate are the current applicable short and long term marginal tax rates. D also includes any gains realized for rebalancing.

Assuming the 529 plan has a wide enough choice such that R529 and Rtaxable are equal, you can tilt things in 529 plan’s favor by choosing a plan with low fees, choosing investments with high distribution rates and be in a high current income bracket.

B. The second condition is that you have to wait long enough for the money in the 529 plans to grow enough to offset the tax penalty.

P529 (1 – MR529) > Ptaxable (1 – MRtaxable)

Where P is the portfolio value; MR529 is 10% + future bracket; MRtaxable is the future long term cap gain rate. Again for things to be in 529 plan’s favor, you have to wait long enough and withdraw the money when your future tax bracket is as low as possible; or to be more exact, when the differential between MR529 and MRtaxable is as small as possible.

Putting the numbers together
Since no one can predict future tax law changes, an exact analysis is not possible. For example, it’s highly debatable whether the current low capital gains and dividend tax rates will be extended after they sun-set in 2010. The best we can do is to take the current tax rates and make reasonable projections into the future. My assumptions are as follows:

Portfolio annual return: 10% (As you’ll see later, I suggest investing with the most aggressive option. At any rate, it matters little when the returns of the 529 plan and the taxable plan are equal.)
Taxable plan fee: 0% (No extra fee besides the intrinsic mutual fund expenses.)
529 plan fee: 0.60% (e.g. the Nebraska program)
Current marginal tax rate: 31% (federal + state)
Current long term capital gains rate: 15% (ignoring any state tax liability here)
Early withdrawal penalty: 10%

Distributions and gains realized for rebalancing as a percent of the portfolio: 6%
– Of which are long term capital gains or dividends: 75%
– Of which are short term capital gains: 25%
TRtaxable: 1.14% (This is the amount of annual taxes to be paid from the taxable account. Since this is greater than the 0.6% expense rate of the 529 plan, condition A is satisfied. Given time the 529 plan will overcome the extra tax burden at liquidation.)

The output is in the table below. For each dollar invested in year 0, I computed the after tax (and penalty) liquidation value of the accounts as a function of the respective future tax rates and number of years invested. Some rows were hidden to make the table smaller.

It can be seen that a LT cap gain of 5% is pretty tough to beat. Even at a low marginal rate of 10% (+10% penalty), it takes 33 years for the 529 plan to catch up ($15.712 vs. $15.695). However, if the future LT cap gain rate goes back to 10%, the break-even occurs after year 20. Still a very long time but doable. The difficulty of the task grows if your future brackets are higher. It will take 36 years for the 15% future income bracket to break-even with a 10% LT cap gain. And if you are fortunate enough to be in the 25% income bracket in retirement, it will take a long, long time (54 years to be exact), to exceed the corresponding 15% LT cap gain.

At this point, you’re probably thinking, “Why bother!” Indeed, 401(k), Roth IRA or the traditional IRA, even the non-deductible kind are much better ways to save for retirement. It is only after those have been maxed out, does the 529 plan emerge as a potential alternative. As described so far, it’s applicable to only a very small segment of the population with high disposable income or a lump sum to invest early on.

If the story ends here, this would not have been a useful exercise. Fortunately, there is much more, both in terms of the 529 vs. taxable plan comparisons and ways of utilizing the 529 for qualified educational expenses thus avoiding the 10% penalty. Please stay tuned for Part 2!

Stages In Life: Retirement Planning

Most people when they finally get out of school and start their first job, are so happy with their financial independence and the first paycheck, that they splurge the entire paycheck, if not more.  Few people think so far ahead about the next stages in life, of getting married, having a family, and eventually retiring from work.  If one doesn’t plan to have children, then they have plenty of time to save for their retirement.  But if they plan to have children, the best time for saving money is really right after schooling is over, and before having any children.

For the ease of this discussion, I’m going to assume some simplistic scenario.  Let’s assume that a person named Kevin graduates from college at the age of 22.  For the next two years, Kevin either spends another 2 years for graduate school studies, or simply lives from paycheck to paycheck because of little work experience which leads to lower salary.  And after age of 24, to age of 30, Kevin develops the career, and gets married.  At about the age of 30, Kevin has his first child.  The first child goes to all the levels of schooling, and graduate from high school after 18 years.  At that time, college expenses will start to kick in when Kevin is age of 48.  Assuming that Kevin has two kids, separate by 3 years in age.  So for the next 7 years, Kevin will need to help out college expenses for his kids.  Once it’s over, Kevin is 55, and has another 10 years of working & saving money until his planned age of retirement at 65.

Now, let’s look back at the timeline of Kevin’s life in regards to earning and saving potential:

  1. From 0 to 24: it’s pretty much zero.  At the end of 24, Kevin may have some or a lot of college debt.
  2. From 24 to 30: this is a time that he can potentially save some money (either towards paying down college debt, or just simply saving).
  3. From 30 to 48: Kevin’s earning may grow with experiences, but his expenses probably go up as kids grow up.  Let’s assume that the increase in earning is offset by the increase in expenses.  Actually in reality, expenses are a lot higher for the beginning years because the wife cannot go out to work.  If the wife goes out working, the preschooling or daycare expenses may take up all of her paycheck.  Either way, you end up with less (retirement) savings per husband and wife.
  4. From 48 to 55: I think we can safely assume that the saving will be zero or negative if Kevin decides to pay the majority of the college expenses for his children.
  5. From 55 to 65: This is another good 10 years of earning and saving, especially with Kevin’s long time work experiences.
  6. From 65 to 90: That’s 25 years of living expenses that need to be saved up (for both husband and wife).

So you want to tell me when Kevin can save for his retirement?  To err on the conservative side, and for simple discussion, let’s assume that Kevin’s investment return is only on-par with inflation rate.  So in terms of buying power, what he save every year will simply be what he can spend later.  Depending on the ratio of Kevin’s annual saving to Kevin’s annual retirement expense, the entire retirement picture can look very different.  If we assume that the sum of Kevin & his wife saving is about the same as their retirement expenses which will probably be more than $30000 every year, we can simply count the number of years of savings to find out how many years they can retire.  And that’s 6 years from 24 to 30, 18 (or probably less) years of savings from 30 to 48, and 10 years from 55 to 65.  That’s total of 34 years of savings for 25 years of retirement.  Boy, I haven’t counted the college debt at the time of graduation, and I forgot about buying a house to live.  But for the house, I would also count the housing equity as part of your saving, more in terms of the mortgage that you have paid down, and less in terms of equity due to the current high housing price.  So you can “keep” part of your saving in the house, and reverse mortgaging your equity at the end.  It’s almost like another bank account.

So with ratio of 1, that’s 34-25 = 9 years extra savings, either for your heir, or for your margin of errors, such as discounting the first 5 year savings when the kids are 0 to 5, or discounting for negative saving years when they’re in college.  Now, if you have spent thru your paycheck from 24 to 30, you have just made your margin of errors to be pretty close to zero.

For singles, the scenario is much better.  There are no kids to raise, nor kids’ college expenses to pay.  All the increase in salary earning can go towards extra saving (& inflation), not counting the extra 7 prime years of age 48 to 55.

As for myself, my current saving rate is about $45K, and my estimated retirement expenses are about $31K, summing from my current budget, plus doubling the travel expenses and tripling the medical insurance and doubling the car insurance, and subtracting out any direct kids’ related expenses.  I also add in property tax.  My saving to retirement expense ratio is slightly better than 1, at almost 1.5.

If you want to put in all the details of your retirement planning, including the inflation rate and investment return, existing asset and liability, you can use my retirement calculator.  You can use it to estimate when you can retire.  It’s what I use personally, and it’s very comprehensive, yet simple.

Bottom line, it’s never too late to start saving now.  It is far better to have some savings set aside already, then to worry about whether you can hit your retirement saving goal in time.