MWeddell

I am trying hard not to make any claim about when market timing will work and when it won't work. One can easily imagine scenarios when, with hindsight, it has worked and when it has not worked. What I am trying to say is that if your asset allocation was properly determined, then recent short-term performance should not cause you to change your asset allocation. I can think of exceptions but that's the typical case. This assumes that one's investment goals and investment horizon haven't materially changed. Also, one should still periodically rebalance to one's target asset allocation.

First of all, let me clarify that none of us is giving investment advice. I'm merely trying to provide broad guidance that may or may not apply in any particular individual's situation. Furthermore, I am not personally an investment professional. Kirk, the whole concept of dollar cost averaging doesn't really apply to when one has a lump sum to invest. If, in your hypothetical, one insists on buying into the equity market gradually, this implies that the rest of the portfolio is left uninvested or invested only in cash. This is certain less than ideal, especially if you want to assume that there is a bull market, i.e. one in which stock prices are rising. In that case, dollar cost averaging won't work as well as investing completely when the lump sum becomes available.

jbo, No, I disagree. Let me illustrate with a recent example. In the middle of March 2000, many broad stock indices hit their all time highs but then started to drop sharply. By the end of March 2000, there were already some people who thought it was a buying opportunity. It was popular during the five preceding years to "buy on the dips," to believe that the general trend in the stock market was heading up so that any dips in stock prices were most likely temporary and were buying opportunities for so-called savvy investors. Increasing the percentage of one's portfolio invested in stocks at the end of March 2000 would not have been a fortuituous decision. Especially with the S&P 500, which is a broad-based large cap U.S. stock index, the efficient market hypothesis is the strongest. If these is a general consensus that the index was likely to head up in the near future, then that expectation would already be reflected in the current price of the stocks in the index. I would be very skeptical of any claims that anyone can predict short-term movements in broad U.S. stock market indices even though one can read these claims all the time. Regard them as entertainment, not information that should impact one's investment strategy.

Kirk, No, I do not intend to imply that one would take a lump sum distribution from the retirement plan. Even if one did take a lump sum distribution, most or all of it presumably would be rolled over. It would be fairly uncommon for someone to sell all retirement savings assets upon retiring (for example, someone who wanted to buy rental properties and live off of the rent during retirement years). Most retirees need to manage their assets for many years after retirement. Given that someone has lived to age 65, their life expectancy is at least 20 more years from that point. However, I still believe that it is prudent to shift some (but by no means all) of one's retirement savings out of stock funds as one approaches retirement. One can quibble whether 10 years was the right rule of thumb, but the general concept is definitely right. First of all, because one is going to start spending one's retirement savings soon, one's investment horizon is becoming shorter and one rationally should start paying more attention to short-term risk. Second, many employees really will experience a significant loss of earning power upon retirement that should make them less able to tolerate short-term losses. If someone retires, experiences signficant short-term investment losses, and then decides to work for a few more years to make up for the losses, the retiree often cannot be reemployed at the same or higher compensation level that he (or she) previously had plus the risk of health problems that restricts one's earning power is growing. Even with a 20 year life expectancy, a retiree understandably should be more adverse to short-term risk and have a smaller percentage of his retirement savings in equities compared to an active employee more than 10 years from retirement. Of course, when one is retired, hopefully one will have a large enough nestegg that it makes sense to purchase individualized investment advice from a fee-for-service financial planner. The guidelines I was proposing probably work better for active employees several years from retirement.

jbo, Buying low is of course a good thing to do. However, to determine at any point in time that one is "buying low" requires knowledge that a fund's value is going to rise in the near-term future. I believe that the typical retirement savings plan participant does not have this knowledge of the future and that in fact it is uncommon for anyone (not in the possession of material nonpublic information) to have that knowledge. Hence, I would not try to buy low. To answer your question more directly, if you see funds going down for 1, 2, or 3 consecutive months, I would not typically suggest that you move money out of those funds. The reason is that any short-term past performance will only very rarely change one's long-term expectation of future performance. If your asset allocation was properly set before the fund started underperforming and your long-term goals and investment horizon haven't changed, then your asset allocation should change just because one of the funds underperformed (other than periodic rebalancing of your portfolio). I continue to advocate that you ignore the noise from co-workers and, for that matter, much of the personal finance literature. If you are using diverified funds, not selecting individual stock, and you have a long investment horizon, then you really don't need to be worrying about your retirement savings investment strategy on such a frequent basis.

I don't think the fact that the profit-sharing contribution was never made would affect it.

If you can tell that a fund is going to keep falling then you are better off to pull your money out of it and if you can tell that the fund has bottomed out and is going to rise then you are better off to move your money into it. Those are awful big ifs, aren't they? Just ignore the noise, jbo. Your co-workers who had 100% of their accounts in the S&P 500 either have enough time to make up their losses or are close enough to retirement that they shouldn't have been 100% invested in equities. I continue to advocate that you ignore recent past performance when setting your asset allocation goals.

Employees who already have the 3 years of vesting service must of course remain 100% vested. Others can be moved to the 2-6 year graded vesting schedule. See Treas. Reg. 1.411(a)-8T. Of course, one should expect some employee dissatisfaction with that approach.

The Company A employees are still in a separate plan, right? Yes, they can be treated separately for ADP testing assuming they also are treated separately for coverage and other testing purposes.

There's no guidance other than the language of the 401(k) regulations themselves. If it is called a school and it requires a high school diploma to attend so that it clearly is "post-secondary," then you probably want to grant the withdrawal request.

I say ignore them both. They are both right but it's just noise that distracts you from your goal of creating and sticking with a healthy retirement savings habit. You are probably better off to just pick a lifecycle fund with a target maturity date close to the date you attain social security retirement age. Let the pros do your asset allocation and don't worry about what your co-workers say and don't worry about short-term fluctuations in your account balance. If you want to make your own asset allocation decision, then some guidelines are Make sure you gradually decrease the percentage of your account invested in stocks beginning about 10 years before you hope to retire because the short-term risk starts to become more of a consideration then. If you are more than 10 years from retirement, then put the vast majority of your account in equities because the expected return is higher than cash or bonds About 20% of equities in international stock funds, 80% in domestic. Of your domestic equities, put the majority in large cap funds. Use an S&P 500 fund for your large cap portion of your portfolio because, there's usually one available in 401(k) plan investment line-ups, it'll prevent you from tilting toward growth or value too badly, many studies suggest that this is more likely to succeed than expecting whatever domestic large cap manager(s) you pick to outperform the index, and because you don't have to worry about monitoring your manager. Balance your small and mid cap domestic equity funds between growth and value. Co-workers who lost too much in 2000-02 probably were picking funds based on recent past returns and ending up with much more money in growth funds than in value funds. Don't invest any money in your employer's stock unless you are forced to do so. You've already got your career invested with that company. Use only diversified funds. Revisit your goals annually. Unless you are within 10 years of retirement, many years all you want to do is rebalance your account but keep the same asset allocation. Occasionally, if your plan offers more than one fund in an asset class, you may want to switch funds but this should be a rarity. Try not to be fickle. Keep on saving! Don't be discouraged by short-term results. Retirement savings is a marathon. That's just my opinion. Free advice is usually worth what you paid for it.

To clarify, one must disaggregate the union and nonunion employees for testing purposes. This assumes that benefits were the subject of good faith collective bargaining and that the union doesn't cover only professional employees. Second point: there is no ACP testing for union employees. I agree with the more general point, that the plan design can be a safe harbor design for the nonunion employees.

Whether the match forfeitures are allocated to all participants or to just those who defer depends entirely on what the plan document says. Either method is permitted by law. If they are allocated to all participants, then they are testing like a nonmatching contribution and not included in the 401(m) test.

I used to be more aggressive on this issue. However, when there isn't a deadline indicated (which I think is the case here despite Bob K.'s post above) and where none of the more specific deadlines applies (e.g. this isn't a remedial amendment period issue for a plan that already specifies a testing method), then the amendment should probably be done within the plan year in which it becomes effective. This is a general principle of tax law, not something where one can easily cite the relevant authority.

I also understand that Roth 401(k) contributions are treated like elective deferrals for purposes of the ADP test.

There are several possible correction methods for the annual additions in excess of the Section 415© limit. However, the plan document must specify which method is used: the plan administrator does not have discretion in this regard. Of the plan documents I've read, the most likely correction method is to refund elective deferrals in excess of the 415© limit, but you'll have to check your own document.

With 200 eligible employees, it seems like the benefits would be worth the administrative hassle, but again it's just my opinion. The most important con is that having a 401(k) with no match might increase the demand for a match. That strikes me as pretty much irrelevant where there's already a 15% profit-sharing contribution. If one is concerned about administration hassle, the HCE limit could be set so that the plan automatically passes using the prior year testing method. At the least, this would let the age 50+ HCEs make catch-up contributions.

If you exclude these NHCEs from both the elective deferral and the matching contribution portion of the plan, then it would be just a coverage testing issue. However, under the facts you presented, Tom has it right (as usual): the matching contribution would not satisfy the safe harbor requirements.

Yes, I have two church clients that use 403(b)(9) accounts. We saw the advantage as being able to obtain a diversified stable value fund instead of a guaranteed account that was guaranteed by just one insurer and that tended to be less liquid if we ever changed providers.

Yes, the rules that were part of the SBJPA of 1996 also were originally proposed by Sen. Boxer.

Yes, there are some rules in ERISA Section 407(b)(2), but they aren't the rules that katieinny is recalling. Those rules permit participants to elect to put 100% of their deferrals into employer real property or employer securities but limit a plan's ability to mandate that elective deferrals be invested in those assets. There's a huge exception for ESOPs. Those rules were added back in 1996 (I recall) after the Color Tile bankruptcy. There weren't any changes to these rules after Enron failed.

There is 401(a)(4) testing of vesting schedules. Technically, it's not BRF testing, but the arithmetic can be the same as the BRF test. Hence, if one of them is an HCE, then yes there may be problems. I don't see a problem though if they are both NHCEs and the employer is aware of the cost implications of doing this.

When you're testing the 403(b) plan, then you may indeed aggregate it with a qualified plan. IRS Notice 89-23 (which will cease to be current law whenever the proposed 403(b) regulations become final) describes this in much greater detail than the Reg. 1.410(b)-7(f). However, when you are testing the qualified plan, you can't aggregate the 403(b) plan. I've never tested it that way but you may be able to run one 401(m) test for the 401(k) plan and a second 401(m) test on the aggregated 401(k) and 403(b) plans. You'll need to go through Notice 89-23 to make sure that the aggregated test is permissible instead of running a separate 401(m) test on just the 403(b) plan.

No, you can't aggregate a 401(m) arrangement within a 403(b) plan with a 401(m) arrangement with a qualified plan for testing purposes. The 401(m) regulations will define "plan" based on a cross-reference to the 410(b) regulations. That definition of "plan" doesn't purport to cover 403(b) plans at all. See the second sentence of Treas. Reg. 1.410(b)-7(a).

It's not very clear. The annual notice requirement is in various revenue rulings and in an IRS General Information Letter issued in 1Q2004. However, the final 401(k) regulations omit this requirement. It's hard to know whether the final 401(k) regulations intentionally omitted the requirement given that the matter wasn't discussed in the supplementary information and given that the 2004 general information letter was released after the 401(k) proposed regulations (which didn't have the requirement either) were issued.