Monday, December 17, 2012

Taking another look at Eaton Corporation

Eaton, ETN, has had a nice run in price since I bought it in July.  I identified ETN in a screen I ran on FAST Graphs shortly after subscribing to the service.   At the time, ETN looked to be priced about 35% under fair value.  I didn't do too much additional diligence, and purchased shares on July 5, 2012.

I purchased additional shares on July 6, July 12, and August 13, and received a dividend which was reinvested on October 19.  My weighted average cost for ETN now sits at $39.95.  ETN trading today at a price of $53.93.  This is a gain, including dividends reinvested, of 34.98%.

My initial premise was that the stock was under fair value by 35%, and now I've gained 35%.  Is it time to sell the position?  Let's take a look at the current valuation vs. earnings.

Here an historical look at earnings and price movement for the past 15 years:

Based on this graph, the price still needs to rise to about $61/share to reach "fair value" based on the average PE given over this 15 year period.

However, most stocks have suffered from "PE Compression" since the great recession of 2008, so I like to do a six year graph to get a more up to date picture.  Here is is for Eaton:

If you look at the blue PE line, you see the average PE for ETN has come down from the 15 year average of 14.9, to a six average of 13.9.  "Fair" value based on this ratio is $57.37.  Thus, today, ETN is trading below fair value, but only by $3.44 or 6%.  Moving to technical analysis (gasp), here is a YTD price chart:

From this you can see that July was a pretty good time to buy, and today may be a decent time to sell.  I'll expand the graph to three years:

This has me wondering about the PE ratios at each of these tops, so I'll go back to FAST Graphs, and check each below:

So, I went back over the black price line, and pulled the prices at each peak and valley, which FAST Graph gives you easily, but then divided by earning listed for that year.  Unfortunately, I can't figure out quarterly earnings for each of these points, and haven't found a source for these.  My PE numbers listed seem incorrect in absolute terms.  Correct directionally for this comparison, but incorrect nevertheless.  If the blue line represents PE=13.9, how can a point above the orange line have a PE of 13.8?  I'll have to ask Chuck.  Anyway, from this chart I conclude that ETN may be reaching fair value, and may be due for a dip after a nice run up from my buy point this summer.

Therefore, I'm placing a 4% trailing stop loss on the shares, which will guarantee me a sell price of at least $51.73 based on today's prices.  Let's hope it keeps climbing before execution.

Best Chump

Portfolio Snapshot December 17, 2012

Below are the latest holdings for my IRA.  Those following along will see that I've held on to ETN and added Kohl's and Omega Healthcare.  I have 28 holdings, and almost no cash.

The stocks are sorted by 5 year estimated total return based on the excellent FastGraph software.

I'll follow this up in a few days with a spreadsheet update showing gains/losses, and % of portfolio.



Thursday, December 6, 2012

Should I Convert my Dividend Growth IRA to a Roth IRA?


**Note to readers:  On Sunday, December 16, an astute Seeking Alpha reader pointed out an issue with my analysis.  After checking my spreadsheets, I determined that he was correct.  In my original analysis, I start removing $30k from the accounts to live on.  I index this number for 3% inflation throughout retirement.  As I take the money from income and RMDs, I correctly tax the withdrawals.  However, as the reader pointed out, I should be taking more pre-tax money from the conventional IRA so that the after tax amount gets me to parity with the amount I withdrawal from the Roth IRA.  So, to get $30,000 for expenses from the conventional IRA, I need to withdrawal approximately $45,600.  This reduces the value of the conventional IRA, and leaves less excess cash for reinvestment into a standard taxable account.

This change in the comparison effects case 1 most strongly;  The conventional IRA is still more valuable up to age 85, but the Roth IRA growth far outpaces the conventional IRA from that point onward.  Case 2 and Case 3 change slightly, but the outcomes and graphs remain very close to the originals.

I've replaced the conventional IRA spreadsheet with a revised version, and the Case 1 table and graph.

My conclusions and summary remain the same, but I thought it important that I show the proper data in case 1.

Sorry for the inconvenience!

While reading through some great articles on Seeking Alpha, the comments turned to conversion of an IRA to a Roth.  I was not aware that the laws had changed to allow creation of a Roth with converted IRA funds.

Unfortunately (or fortunately), our "married filing jointly" income exceeds the IRS limits for contribution to a Roth, so we don't currently have one.  Over the years, having worked at several large firms, I've been rolling my 401k plans into a conventional IRA, which is growing nicely.  After learning that a conversion was now possible, and reading some helpful comments on SA,  I decided to consult with my accountant.  Here is the gist of what I learned, though please keep in mind, I am far from expert regarding IRAs and tax rules.

In 2010, congress made it possible for those previously not eligible for a Roth IRA, to convert IRA proceeds to a Roth.  Due to income restrictions, I cannot openly contribute to a Roth, but I can convert a portion or all of my conventional IRA to a Roth.  The catch, of course, is taxes.  My IRA was funded entirely with pre-tax earnings from 401k contributions, thus, I need to pay federal and state income tax on any portion I choose to convert.

There are many attractive features of a Roth IRA, and I'd like to have one (whoa, I sound like one of my children).  They aren't taxed when you withdrawal money after age 59 1/2, and you aren't forced to take money out starting at age 70 1/2, the dreaded "required minimum distribution" or RMD in a conventional IRA. After you die, your Roth IRA can be passed along to your heirs without a tax hit.

Back to the question of should I convert my IRA?  The answer is far from simple.  There are many factors, and modeling them all (which I've attempted to do) is pretty cumbersome.  Age today, age when you plan to retire, income needs in retirement, tax bracket today, tax bracket with proceeds you are converting, tax rate when you retire, % of IRA funded with non deductible contributions; well, you get the picture.  I've found some calculators online and played with them, but I don't know the assumptions used and get differing numbers on different sites.  Thus, I decided to build a spreadsheet and model some scenarios myself.


Changing numbers to protect my privacy, let's say I've managed to accumulate a conventional IRA worth around $200k at age 48.  I'm going to look at three different options for conversion:
  1. Convert $15k per year until I retire at age 65
  2. Convert $100k per year over two years
  3. Convert $100k in one year, then stop
Taxes on the converted amounts will follow my current marginal tax rates of 33% federal, and an onerous 9.3% state (in my sunny state).  Care needs to be taken with the amount converted in any one year that the income doesn't put you into a new higher tax bracket, further damaging returns.  The schedule below shows today's federal tax brackets:

I'll be comparing these four options to my conventional IRA, which, at age 70.5, will require minimum distributions every year.  I'll assume a reduced marginal federal tax rate of 25% and the same state rate of 9.3% for these distributions, and will re-invest them into a standard taxable brokerage account.  For you lucky dogs living in low tax red states, I'll also look at a comparison with no state taxes.

I'll assume that both my current conventional IRA, and the taxable account I create with the RMDs after age 70, will have the same pre-tax performance; 6% growth in share price, 3% portfolio yield, and of course, dividends will grow every year by 6%.

First, a look at my projections for the conventional IRA:

Conventional IRA for Comparison

To keep it simple, I'm not adding any money to the IRA.  Further, dividends are reinvested until age 70, then used to fund a new taxable account with the "required minimum distributions."  At age 65, I start withdrawing cash for living expenses.  I assume I'll need $30,000 growing 3% per year to cover inflation.   However, I need after tax dollars, so I need to withdrawal more when you consider federal and state taxation.  The "Pre Tax Needed" and "Shortfall or Excess" columns do this calculation and shows how much is either needed via sales of shares, or in excess available for reinvestment.  I realize this is not enough to live on, but for this analysis, I'm assuming I will have additional sources of income for retirement expenses.

At age 70.5, I need to start pulling out the "required minimum distribution" or RMD.  The formula for required minimum distribution is based on the number of years the government thinks you will continue to live.  You take your assets in the IRA on December 31, then divide by the number of years you have left according to the handy table below:

Predicting how long I have left is bit depressing, as is watching the RMDs bite into my hard earned assets.  But don't fret, I'm not taking the RMD distributions and burning the cash in the fireplace; I'm reading Seeking Alpha, and investing the proceeds into...drum roll...what else, but a dividend growth portfolio of quality stocks.

From the RMD proceeds at age 70 1/2 and beyond, is born the new taxable dividend growth portfolio.  To further complicate matters, only income from IRA dividends can be reinvested into the IRA, while excess RMD cannot (don't tax rules make life fun?).  So, excess RMD leaves the fund and goes into the new taxable account, and excess income from IRA dividends, in excess of the RMD and what I withdrawal to live on, stay in the fund for reinvestment, and are not subject to taxation.  Using the same continued assumptions for growth, but now using after tax (federal + state) RMD and after tax dividends for reinvestment, I constructed a taxable dividend growth account similar to the IRA, it's shown below:

Taxable Account from RMD Excess

So while the original IRA is required to distribute an increasingly large chunk of cash, our DGI portfolio, even after taxes, helps soften the blow.  Please note, I increased the tax rate for this portfolio as I moved into new tax brackets based on income received from RMDs and dividends.

Now to the Roth conversion.  For this part of the analysis, I take the same IRA as in the first chart above to start, then pull uniform distributions out every year.  Per my assumptions, I'll take the $15,000 per year, pay taxes on these funds, and create a Roth IRA based on dividend growth stocks (of course).   For purposes of these analyses, I'm going to pay the taxes with money held elsewhere in a taxable account, to allow more money into the Roth and avoid any penalties while under the age of 59.5.  So to recap, I'm comparing my IRA + a taxable account that starts with $100,000 in it and grows by the excess RMD payments in retirement versus a conventional IRA with funds pulled out until retirement for conversion into a Roth IRA + a taxable account with $100,000 from which I will pay the necessary income taxes during the conversion years.  Wow, just explaining the planned analysis gets confusing!

Next is the IRA from which I'm converting funds:

Case 1 IRA less Converted Funds

Interestingly, even converting $15,000 each year until age 65, the portfolio manages to grow modestly.  After the conversions stop at age 65, the IRA grows more rapidly until RMDs kick in at age 70.5.  Conversions out of this account and into the Roth IRA are fully taxed, both Federal & State, then grow and re-invest tax free forever, with no taxes due when the proceeds are withdrawn.  The associated Roth IRA that I've created from the above conventional IRA looks like this:

Roth IRA Built with Converted Funds

At age 65, I start withdrawing $30,000 per year, growing 3% per year, to help with living expenses.  And because this is a Roth, I get the money tax free!  Further, because there is still IRA money left growing, I'll need to pay RMDs again at age 70.5.  Those monies will go back into the taxable account from which I paid the IRA conversion taxes.  The $100k taxable account from which taxes were paid, and into which RMD payments are going, is shown below:

$100k Taxable Account Tied to Roth

Analysis Results

The table and chart below summarize the results of the first analysis, case 1.  Total account value and annual income for the two account types are presented at different ages:

Case 1 Results

Here are the same data in a simple graph:

Case 1 Graph

Up until I hit age 85, the conventional IRA retains a greater overall value, and provides more income annually.  After age 90, the Roth catches up and passes the conventional IRA quickly.  From the analysis, you can see the huge negative effect of taxes.  The $15,000 per year conversion is taxed at a combined rate of 42.3%.  The tax bite is too much to overcome until age 90 in this scenario.  Running exactly the same analysis in a state with no income taxes, the situation for the Roth is improved slightly, with the Roth IRA overtaking the conventional IRA around age 85, or five years sooner.  Even with no state taxes, the analysis favors the conventional IRA until age 85, though the difference in the outcomes is reduced slightly.

In the second analysis, case 2, I convert money more quickly.  $100k per year over two years.  The tax hit will be high, but the money in the Roth IRA will have more time to grow.  Here is the case 2 data summary:

Case 2 Results
And the associated graph of the data:

Case 2 Graph
The conventional IRA is a clear winner.  The large tax bite paid over two years, $84,600, virtually killed that taxable account compared to an equivalent taxable account where no taxes are paid for conversion.  In the conventional IRA case, that $100k taxable account, even after paying income taxes every year, grows to a whopping $355k by age 65, and over $500k by age 70.  Using taxable account savings to pay the conversion taxes hurt the outcome significantly in this case.

For my third case, I'm converting $100k in the first year, then stopping the conversions, and letting both the original IRA and converted Roth grow.  Here is the summary table:

Case 3 Results

And the associated graph of the data:

Case 3 Graph

In case 3, where $100k is converted in a single year, the results between the two accounts are much  closer.  The IRA still provides more total value and income throughout the analysis, but by a small margin.  One cautionary note;  $100k converted in a single year (both case 2 and case 3) could put a large chunk of your income into a higher tax bracket, and thus further hurt returns for these cases.  I would likely convert exactly the amount necessary to get me to my next rung on the income tax table, and no more.

The other big factor in all of these cases is your State of residence.  If you live in a low, or no tax red state, the analysis gets more favorable for a conversion.  As in case 1, I repeated this analysis for 0% state taxes vs. the 9.3% in the graph above, and the results get even closer to even.   The graph for the 0% state tax is below:

Thoughts and Conclusions
  • At my current age of 48, keeping my money in a tax deferred IRA seems a better choice than converting any or all to a Roth IRA.
  • The best result came from case 3, converting only 1/2 of my IRA, and then heading into retirement with three accounts;  a conventional IRA, a Roth IRA, and a taxable dividend growth account.
  • At age 65, the converted Roth IRA fund value and income will always be lower than the IRA due to the taxes paid upon converting.
  • The Roth IRA really "shines" vs. the conventional IRA once you are retired and withdrawing funds.  Less money comes out, and more is available for reinvestment vs. the conventional IRA, due to taxes.
  • High state income taxes hurt the argument for conversion.
  • Minimizing tax loss when converting, and preserving capital with time for growth are the most relevant factors.
  • Paying taxes out of the funds you are converting is a bad idea.  If you are younger than 59.5, a 10% penalty is due in addition to state and federal taxes.
  • Our government seems to be the biggest beneficiary of a conversion.
The decision to convert funds from an existing conventional IRA to a Roth IRA are effected by myriad variables.  Age, tax rates, amount planned for conversion, income required in retirement, and how long you "plan" to live, all play a big role in the analysis.  From the "not so simple" analysis above, I'm leaning toward doing nothing.  Just sticking with my current IRA and taxable stock account.  That said, the thing I like about case three, or converting a portion of my IRA to a Roth IRA, is tax policy diversification.  It seems reasonable to assume we can't know what changes to tax policy we'll see in the next thirty years or more.  Having three types of retirement accounts versus two might be beneficial depending on how our laws change in the future.