Self Directed Retirement

I’ve read more than a few passing mentions of so-called self directed retirement accounts.  I never really understood what that meant.  But the message was usually delivered with lots of exclamation points:

  • As a self employed business owner, you can reduce your taxable income, save for retirement, and increase your capital for investment!
  • As a full time employee with a side hustle, you can direct your retirement savings into your side business!

See!?  Why are you still reading these words?  Who wouldn’t want to have a self directed retirement account?  Why haven’t you called a broker yet?


Not so fast. Let’s dig into this a bit further.

Disclaimer.  Keep in mind: I’ve never formally studied retirement planning or financial advising.  Any decisions you make are your own. 

Let’s start with my motivation: with a potential opportunity to acquire our second (and maybe third!) on the horizon and a need for more cash, I figured it was time to figure out how we could afford our next opportunity.  And, a self-directed IRA represents a way to get cash without selling a kidney.  Here’s what I found:



What is a self directed IRA? 

It is a tax-advantaged (tax deferred) account offered by the US Government (IRS).  The account requires a custodian and allows for investment into alternatives (e.g., real estate, notes, or private companies). The account is funded with pre (traditional-IRA) or post (Roth-IRA) tax dollars.  The account then grows or shrinks as your investment choices grow or shrink and is subject to similar tax rules with a few extras.  For example, a self-directed IRA is prohibited from a bunch of transactions with “disqualified persons,” namely anyone closely related to you or businesses where “disqualified persons” have more than a 50% stake.

Honestly, It’s a bit tough to find details on the IRS’s website in one place (odd, right?).  I found that the fine folks at Cornell Law publish the internal revenue code  from which all the retirement accounts are authorized.  And, there’s an innumerable set of documents that the IRS publishes on all manner of arcane tax rules.  I also found IRC 4975(c)(1) and IRS Publication 590 to be the most referenced ones.

I did find a bunch of company websites that are selling their services as custodians.  Note: I’m not affiliated with any of these companies, nor am I receiving any compensation for mentioning them.  Some of these have pretty thorough documentation of the ins/outs of a self-directed account.

OK.  Enough overview.  I want to buy real-estate with my retirement dollars already.



Why is a self-directed IRA useful?

You can invest your retirement dollars into a wider variety of investment types.

Typical retirement accounts (401(k), 403(b), IRA, Roth IRA) are set up through an investment company like Vanguard, Fidelity, etc.  In a typical account, you can choose to put your money into stocks, bonds, mutual funds or other publicly trade-able assets.

I don’t want to do that. I want to buy a building or a note or shares of Tesla before they went public, and I want to do it with my retirement dollars.  This site lists a bunch of different types of assets (some of which you can invest in through traditional IRAs):

    • Real Estate
    • Private Company Stock
    • Tax Liens & Deeds
    • Oil, Gas & Mineral Rights
    • Crowdfunded Ventures
    • Trust Deeds & Mortgages
    • Private Loans to Businesses or Individuals
    • Venture Capital
    • Precious Metals
    • Traditional Stocks, Bonds & Funds
    • Anything the IRS rules allow for

Here’s one of the exciting benefits of using retirement dollars this way: your self directed IRA can get a non-recourse loan to finance the purchase of a property.  What!?  Well, yes.  You have to find a lender comfortable with this, and at typical loan to value ratio is closer to 50% but, it sure looks like you can leverage your retirement dollars.

One of the biggest reasons to look into a self directed IRA is if you are self-employed.  Lets say I decide to “retire” early from my day job to manage my real estate portfolio.  The self-directed approach allows me to continue putting aside retirement dollars, and I could elect to simultaneously supercharge my real-estate holdings.

There’s a few different flavors of self directed accounts: a solo 401(k), a self directed LLC (typically with “checkbook control”), or a business funding IRA.  I’ll leave the detailed overview to you to click through the links.  Suffice to say that each has minor differences in rules and benefits.

How to get started?

Think of setting up a self-directed IRA as starting a business.  That looks closer to the level of complexity to expect.
First, you need a custodian company that does this sort of thing.  Vanguard, Fidelity, Charles Schwab aren’t the go-to places for this kind of investment.  Some of the linked companies in this post are a good place to start your research.  Here they are:

How to fund a self-directed account?

Once you have your account set up, there are two basic ways to add funds into the account

  1. You can convert an existing retirement account into a self directed one.  This will of course require the support of your custodian company. And, all the IRS rules about taking possession of the funds apply.
  2.  Once the account is set up, you can start contributing much as you would a regular IRA.  The specifics will vary depending on which account custodian you’ve chosen.

How do you get your money out?

The short answer is just like any other retirement account: you cannot.   The longer answer is that you cannot until you reach the minimum age to begin distributions without paying a penalty (all the usual IRS rules apply regarding early withdrawals).

So you’ve invested in a nice little rental property.  You’re earning a great return on your investment.  What do you do with the proceeds?  You can either re-invest in the existing property (e.g., pay down any non-recourse loan balance) or save the cash to buy your next property!  Either way, don’t just take a distribution without ensuring you understand the rules or you may end up paying a penalty and income taxes.

Some other pitfalls of a self-directed account:

  • Cannot use money to interact with “disqualified parties.”  For reference, here’s the list of folks who cannot interact with the funds in the plan:
    • The IRA owner (don’t think you’ll buy a beach rental for your family this way!)
    • The IRA owner’s spouse
    • Ancestors (Mom, Dad, Grandparents)
    • Lineal Descendents (daughters, sons, grandchildren)
    • Spouses of Lineal Descendents (son or daughter-in-law)
    • Investment advisors
    • Fiduciaries – those providing services to the plan
    • Any business entity i.e., LLC, Corp, Trust or Partnership in which any of the disqualified persons mentioned above has a 50% or greater interest.
  • You likely won’t get any advice from the custodian on the tax implications, investment soundness, or legality of any moves you make.  Make sure you know what you’re doing or that you’ve hired a team who can give you good advice!
  • Variable costs.  I’ve not seen many published values, but Clint Coons suggests setup fees run $1500-$2000 and $300-$700 annually.  Depending on your asset base, a self-direct IRA will likely cost more than typical IRAs that limit your investment options to stocks and bonds.


What am I planning to do?

For now I’m focusing on more traditional approaches:

  1. I don’t have enough cash in my non-401(k) retirement accounts to justify the transaction costs of a self-directed IRA…and I’m not quite ready to quit my day job!
  2. I still believe in diversification, and real estate makes up a slightly out sized portion of our total net worth.

Presidential “Pop”


An upset worthy of hyperbole: Donald Trump will be our next president.

Almost immediately, the markets started reacting.  US Futures dropped like a rock.  Foreign markets dropped for real and then rebounded.  Now, almost a week later, lots of folks are starting to interpret what a Trump presidency means.  I wondered what typically happens shortly after an election.  Let’s just stick with the S&P 500 and look at how it closed on November 1 (before an election) vs. December 1 (after an election).  Maybe we can see how panicked or calm we should be…

Here’s a table showing the 1 month closing values and the % changes since 1900.  If you’re wondering why that particular window, it covers a pretty significant span of US history and a range of tumultuous to halcyon periods.  By the way, the S&P Data is from here.   Special thanks to Robert Shiller for making it freely available.  And, the Electoral College data is from here.

One month S & P 500 changes since 1900.

One month S & P 500 changes since 1900.


For all years since 1900 without a presidential election, the average one month change was a positive increase of 0.5% (standard deviation of 0035).  For all election years in the same period, the one month change was a 0.3% increase (standard deviation of 0033).  Here’s a histogram to help visualize things.  See the overlap?  The stats seem to suggest that for this time window, there’s no net benefit to either jumping into or out of the stock market in response to election results.



But, wait a minute.  Aren’t Republican presidential candidates better for business and the stock market?  In the days since The Donald was elected, the US stock market hit record highs.   See!  Well, let’s look at the data.  For all election years in our data set, we can split the one month returns by the winning party.  When Republicans win the White House or Democrats win the White House, there is no difference compared to non election years.

So, forget about non election years: let’s pit Democrats against Republicans directly.  Since 1900, there have been 15 Republican presidential terms and 14 Democratic ones.  Here’s the histogram of the one month change in the S & P 500.  Visually, it looks like there might be something there: a couple more high returning years in red vs. blue.  Again, the stats don’t show any meaningful difference between the two parties.



So, what’s the take-away?  Warren Buffet’s sentiment: “For 240 years it’s been a terrible mistake to bet against America, and now is no time to start.”  Despite all the rhetoric, punditry, and discord, America is a great nation to belong to. 


Regardless of your political affiliations, invest.  Invest for the long haul.  Plant that seedling as soon as you can.  It will grow into your own money tree.

I Have a Dollar. What Should I Do With It?

George Zoomed

It depends.

Of course.  Smart A$$.

No, really. It depends on a ton of different individual factors that are unique to your specific situation.  Funny enough, I’ve been taking about this very question with a bunch of folks. One friend has retirement savings that need a new purpose. Another is debating between adding debt to purchase a rental property vs. pay down their home mortgage.  And, we’re asking a similar question about how we could move closer to our work locations and acquire a second rental property.  My point is simple: everyone faces these kind of trade off decisions routinely.  I don’t think I could or should answer the question for anyone else, but I think there is a framework that folks can apply to their own situation.

At the heart of this topic is the concept of opportunity cost.  (Hold on to something solid; this gets heavy quickly.)  Opportunity cost is what you give up when you commit.  Just got married?  Congratulations; you just gave up your chance to date Claudia Schiffer (Or Justin Timberlake).

Bought a CD?  For the money you invest in CDs, you’re giving up on the chance to buy Amazon stock with those dollars.

Bought Amazon Stock?  Those dollars won’t be buying into the Uber IPO.

Bought Uber?  Those dollars won’t be buying a rental property for you.

To get more technical:

…opportunity costs are used to measure the differences in returns between a chosen investment and one that is forgone. For example, consider a person who invests in a stock that returns a paltry 2% over the year. By placing his money in the stock, the investor gives up the opportunity to invest in another investment, such as a risk-free government bond yielding 6%. In this situation, the opportunity cost is 4%, or 6% – 2%.

Read more: Opportunity Cost Definition | Investopedia

Follow us: Investopedia on Facebook

Sounds easy doesn’t it?  Let’s take a more personally  relevant example to understand some of the nuances of this kind of decision.  Say that I have an extra  $1000 sitting around.  As cash, it might as well be under Aunt Bertha’s mattress.  What should I do with it?  I see a couple of options.

  1. Pay down bad debt: one of our car loans
  2. Pay off our other car loan and start a “debt snowball
  3. Save/invest it for our next rental property

Option 1

This one is pretty straightforward.  Say our loan balance is about $5000.  The interest rate is 2.25%.  We’re a few months into this one due to the untimely demise of my beloved Scion xB.  Of course, our first car loan doesn’t change at all.  So, by paying down Loan #2,  we payoff the loan 14 months earlier, and avoid about $80 in interest over the life of the two loans. Our total interest paid for Option 1 would be $743.02.

Option 2

After almost 4 years, we have about $1000 left on our first car loan.  Let’s say our payments are $100/mo.  So, by paying the loan off now, we save a whopping $11 in remaining interest and free up $100/ mo to apply to the other car loan.  This is the start of the “debt snowball” popularized by folks like Dave Ramsey.  In short, it’s an emotionally satisfying way to pay down your debt.  Although, it’s not financially optimal(this is an awesome tool; check it out for your specific situation).  After paying off the first and applying the payments towards the second, we  find the total interest paid for Option 2 is $689.60.  Sounds a bit better.

Option 3 

This is the most complex of the three. We carry the existing two loans without avoiding any interest.  The rates are pretty low, so maybe it’s OK.  Now, a low cost bond fund has historically returned 3% (Vanguard’s has been a bit better with all the craziness going on over the past few years).  We’re looking to invest the money for three years.  If we assume past performance is indicative of future (possibly a BAD assumption), we’re looking at a potentially life changing $91.32 in distributions.  Our net interest paid for Option 3 would be $651.70.

So, how do these three options compare?  Someone play the “wah, wah” trumpets.  OK, I’m hoping that while our numbers aren’t terribly exciting, this concept is.

Opportunity Cost

Trade-off between paying down one loan, debt snowballing another, and investing funds instead.

It looks like Option 3 provides us with the best overall benefit.  What if we were willing to wait a few more years? What if we consolidated our car loans into a lower cost HELOC?  What if …?  And, that’s why I said, “it depends” way up front.  Your specific situation will vary.  Maybe you have higher rates or larger balances.  Maybe  you’re questioning weather or not you can afford to make extra payments each month.

The point is, you can use math to help make the best decisions possible with your limited dollars.  Those dollars can be the seeds of your future.

Let’s Go to the Casino!

We’re going to take a trip through the Monte Carlo.

Photo Credit: Jamie Adams

What is Monte Carlo?

Nope, not the one in Monaco with the beautiful beaches, fast cars, and the roulette tables.

Not the one in Las Vegas either.

I’m taking about a statistical technique for simulating a process or phenomenon.  Remember that infernal “Bell curve” from school?  It looks kind of like this:

Photo Credit: Jeremy Kemp

All the Bell or Gaussian distribution does is tell you how likely an event is to occur (think of the vertical axis as probability). When you have a known, stable distribution, you can use that distribution to quickly simulate a whole bunch of events.

For example, it takes me an average of 30 mins to get to work.  Some days, it’s shorter. Some days it’s longer.  Once, a car, 3 cars up from me got T-boned while going almost 60 miles per hour.  It took me almost 55 minutes to get to work that day (and no one was seriously injured).  Do here’s what my commute looks like as a distribution:

My commute duration averages 30 minutes. There’s variation. It’s rare for me to make it in 15 minutes or take more than an hour.

Once we know the distribution of historical values, we can use this information to make some educated guesses about what the future will be like.  Remember, when I looked at the long run returns for CDs, Bonds, and Stocks?  All I did was compare actual annual returns against a medium term average: 7%.

Let’s see what a simple Monte Carlo simulation can tell us about the range of future possibilities for our investment.  As always, this is just a thought experiment.  Any investment decisions you make are your own.

Here’s the histogram showing the past annual S&P returns since 1928:

Folks often tout an average S&P500 return of 7%.  Without understanding variation around the average, it’s easy to be mislead.

Here’s the scenario.  At 20 years of age, your Great Aunt Bessie passed away. She hated stocks, bonds, and even CDs.  “Too much risk for me!” She used to chortle from her rocker. “I keep my money cash and close at hand in case I need it.”  So she left you $1000 of cash out of her under-mattress stash. Let’s assume a 40 year investing horizon.  Let’s also assume that inflation and investing costs are zero!  Is Bessie right?  How much money might you have left if you invest it in the S & P 500?

I’m going to use Excel to do the Monte Carlo simulation.  I’m too cheap to spend money on a fancy software package.  And, this way you can do this on your own without needing fancy software.  Using the long run average return might be misleading, so we’ll stick with more conventional short run data 2006-2015.  The average return for the S &P 500 in this period was 9%.  The amount of variation around that average is described using the standard deviation.  In this case it is 0.187.

Don’t freak out if you’ve never loved statistics (most normal humans don’t).  All we’re doing is describing/summarizing the shape of the distribution. In Excel, we’re going to create a column called year.  Start at year 1 and count up to 40. In the next column, were going to enter a formula to calculate the new balance (warning for the faint of heart: there will be an equals sign coming up shortly). Here’s what the setup looks like thus far:

Excel monte carlo setup1

In cell B6, The formula is:


Plug this into Excel so it looks like this (if you’re really scared of doing this yourself in Excel, you can get to a Google docs version here):

In cell B7 copy the previous formula and make one change.  Switch $B$1 to B6.  It should look like this:


Now drag the formula down to year 40.  You just ran a single simulation of 40 years worth of S&P 500 returns.  That’s about all there is to it.  Keep in mind that your actual values will be different from mine, and they will recalculate every time you make a change to the sheet.  It should look something like this:

Excel monte carlo setup2

Now, are you ready to get really nuts!? Copy/drag the second column out 500 times.  You’ve created a (crude) 500 run simulation of the stock market.  Feel that mathematical prowess coursing through your veins?  This is one of the most useful tools you can get for doing some pretty powerful simulations…and it’s almost free!

Let’s look at the results. Copy the last row (year 40) for all runs and then paste values/transpose the data to a new worksheet.

Excel monte carlo setup 3

Once you have a column of final values, you can re-run the data analysis tool to create a histogram of your results.  Again, your specific results will vary, but from what I saw:

Excel monte carlo results

  • In 3 out of 500 runs, you did lose money.  That is roughly a 0.6 % chance.
  • In 151 out of 500 runs, you ended with between $100 and $1000. That is ~30% chance.
  • In 123 out of 500 runs, you ended with between $1001 and $2000.  That is ~25% chance.
  • In 223 out of 500 runs, you ended with between $2001 and %50,000.  That is a ~45% chance.

Tell Aunt Bessie, “Thank you for the gift of money, but you’ll use math to help make decisions about what to do with it.”  Your odds are better in the market than her 100% certainty of gaining nothing.  We’ll keep using tools like Monte Carlo to help make better decisions about a bunch of upcoming facets of our financial lives.  In the meantime, play around with this approach.  It’s a powerful tool for making better decisions.