If you download the spreadsheet I posted, you can shift the assumptions to monthly contributions. The end result will be similar – you get a very high IRR (internal rate of return), because you don’t have to put all the money up front, but make contributions every month. So that last check you write, one month before you cash out, has a very high rate of return since you get 15% for just 1 month of investment.

Personally, I always contributed the maximum to ESPP when I worked for a company that had it.

I have a question for you regarding my specific ESPP. Here are the details:

- I can contribute up to 15% of my base salary to the ESPP;
- Stock is purchased on the last day of EVERY MONTH, at the high-low AVERAGE on THAT DAY.
- I receive the same 15% discount on stock purchased as you describe in your article.
- Any Dividends paid out during the month-long period are reinvested a the time of purchase at the same 15% discount.

I am currently investing 5% of my base salary in my ESPP… based on these details, how advantageous would it be for me to invest more?

Thanks!!

Dave

Your assumption is correct. When I adjust my spreadsheet for a 3 month time period, the IRR jumps to over 300%. Getting a 15% return in just 3 months is huge, so the effects are magnified exponentially from the 6 month case.

I haven’t run the calculations based on taxes, because there are so many complexities when it comes to tax code. Once you let the investment ride with the ups & downs of the market, it’s a very complicated analysis to figure out, on average, whether you are better or worse off, risk adjusted, waiting for the lower tax rate.

If you leave the money tied up for an extra 12 months, then apples-to-apples, the IRR will be much lower. Part of the reason the IRR appears so high is because that last paycheck contribution is only held up for two weeks.

Hope this helps.
Adam

Thanks!

Thats a 10% raise right?

This is just off the top of my head. (1) and (2) are the most common. My analysis really assumes a flat stock, where the delta between the time you get the shares and the time when you can sell has no effect on price.

That is, your first dollar had to wait 6 months in order to get a 15% discount, and your last dollar contributed only had to wait 1 day in order to qualify for the 15% discount. If you could only control one of these dollars, which would it be? If you think about it, you’d want to own the last dollar, since you could have been investing it in other things for the prior 6 months, earning additional interest.

Time has value when it comes to investments, so the fact that your contributions are staggered evenly throughout a 6 month period makes each contribution more valuable, as each contribution to the ESPP boosts the overall rate of return you’ll experience.

Another way to think of it is to assume you could make the entire contribution of $X,XXX on the last day of the 6 month window. That would be the best return, since you could make the same gain overall while only tying up your money for a single day — awesome ROI….but, same $ return you say? Not if you were investing all that money in another asset up to the point you contributed it in the ESPP.

You make some valid points, but in the end, IRR is the only way to value the time-value of money, and in this case, the time-value matters a lot.

If you re-read my post, I do, in fact, state that 17.65% is the number that you might expect with a purely cash-on-cash basis. You’ve done two sequential periods to get an annual number, and that’s fine.

However, it matters that you don’t put all the cash up front. If you had the choice to put all the cash up front, vs. leaving some of the investment to the last two weeks, you would take it. And that’s because, in reality, this isn’t a single investment, it’s 13 sequential investments.

IRR is the correct mathematical way to represent this, so I stand by the 98.4%. Your point is of course valid, however, to illustrate to people reading this article why IRR is not the same thing as a total return.

Adam

I just wanted to add a counter-point to this article. It’s a few days (years?) late, but Google never sleeps.

I consider it a bit disingenuous to use “IRR” in this situation. If you consider cash flows for the period in question (1yr), every “investment” made into the ESPP is getting a literal 17.65% cash return.

After the 6mo’s is up, each penny invested is now worth 1.1765 of itself* (*plus interest). After the end of 1 year, every penny is still only worth 1.1765+interest. Timing does not apply because you have to consider each investment as the start of a single 1-year period.

It doesn’t matter that my last paycheck gets its 17.65% in only two weeks; because after 1 year’s time, the actual return is still only 17.65%+interest.

Let’s do a true example year. In a 1-year period (starting from date of first deduction), you will have invested 4250*2=$8500. The first return gives you $750, plus 6mo’s interest (5%/2) on the whole balance. The second return gives you 750. That’s total cash-flows of 10,125.

So, $1625 in new cash earned in a 1-yr period. With 8500 used to get that, that’s 19.12% actual annual return.

This is the actual, valid, annual rate of return that someone can expect in this ESPP situation. This is what people should consider when deciding to invest in their ESPP. After 1 year (or better, 2 years, the usual offering period) of investing in their ESPP, they will still only be getting an actual annual return of 15%-20% (plus price volatility gains).

- DaftShadow

I am not 100% sure, but I believe that ESPP programs are allocated out of a pool of stock already owned by the company, so there is actually no market transaction, just a transfer of ownership. They use the market price at end of day as an arms-length value to ensure fair pricing.

So there is no possibility of disconnect.

Adam

Thanks for publishing this analysis it has been an eye opener. I have two variations on the theme and would be grateful if you could analyze them.

First, my company’s ESPP cycle is 3 months (not 6). My guess is, this would result in an even greater IRR because the money is tied up for a shorter period of time – true?

Second, I wonder about a variation that takes into account the fact that taxes on gains will be lower if stock is held at least one year. Suppose I always sell after 366 days. This means that during the first year, I tie up my money longer, but in subsequent cycles, I basically keep selling as much as I buy. Could this mean I will suffer a lower IRR only in the first year (i.e. initial investment), but will attain 98% thereafter? Or does it mean that, realizing I would have 9-12 month’s worth of money tied up at any point in time (in the above mentioned 3-month ESPP cycle) I would screw up the whole IRR scheme by trying to save on taxes?

Thanks,

Dan

The reason you are “dumbfounded” is because you are misunderstanding the meaning of an IRR in the first place. An IRR of 98.4% does not mean the money doubled. The IRR is a blended return of the entire cash flow, and is very sensitive to the dates of each investment.

Look at it this way. For the $326.92 that you invested on January 14, it took almost 6 months to earn 15%. If you can earn 15% on your money every 6 months, that works out to an annual IRR of 32.25% (1.15^2 – 1).

But, you also earned 15% on the $326.92 that you invested on June 17th. In that case, you earned 15% in just 2 weeks! If you annualize that IRR, then you get 3686% (1.15^26-1).

When you annualize IRRs, compounding really matters.

The XIRR function is able to calculate the result of a multi-term polynomial that represents all 13 of the investments, plus the single positive outcome, in a mathematically correct blended IRR for the entire stream. That’s why you get 98.4%.

It does not mean your money doubled.

Note – if you play with the dates on the spreadsheet, you will get different numbers if you change any of the inputs by even 1 day. For example, if I make the last payment on 6/30 vs. 7/1, you get a lower number.

The point of this post was to illustrate the correct math, versus the math of the blog I was reading that day.

I have solved enough large, non-linear polynomial equations in my time to have faith in the XIRR function in Excel, when used properly. I also frequently use a calculator to multiply numbers, rather than doing it by hand.

If you are interested in this math, you should research online the basic mathematics of calculating returns on cash flows.

Adam

The Analysis Toolpak comes with Excel – you just have to enable it in the Add-Ons menu.

In any case, IRR & XIRR are totally different functions, so my thoughts are that you are using the wrong function in this case.

Solving for IRR is a very complex mathematical equation for anything more than a trivial cash flow. In this case, you have 14 cash amounts and dates to deal with. IRR makes a number of trivial assumptions to reach its conclusion, mainly by ignoring the date entirely, and assuming the sequence happens over equidistant periods, and ignores annualization.

XIRR uses an interative approach to solve the non-linear equation, using each date specifically for the cash flow. It is the right function to use in this case.

My guess is that besides using IRR, which is incorrect in this case, you also may be providing a poor “guess” to the function (the second parameter). These equations tend to have multiple answers/solutions. (in fact, a polynomial with 15 terms potentially may have 14 different zeroes), so with a default guess of zero, you are getting 2.3%.

Hope this helps.
Adam

Because of ongoing confusion caused by Mark’s comment, I’ve updated the post with a link to an Excel spreadsheet that I used to generate the original numbers for the post.

The 98.4% number comes from the XIRR function in Excel, which is utilized to calculate the correct IRR for a set of non-periodic cashflows.

Thanks.
Adam

You are completely wrong. The IRR is 98.4% in the case I cited above. I think you are confusing the definition of IRR, which is an internal rate of return based on the actual cash flow time periods and an annual ROI, or annual return on investment.

It does not matter that the money you put in an ESPP does not compound or earn interest. Every paycheck, you make an “investment” in your ESPP, with a single large payout at the end.

As a result, the implicit “return” on each paycheck is different from an IRR perspective. The first paycheck sits in the account for 6 months to get its 15% discount, the last paycheck only sits for 2 weeks.

The fact that you can’t re-invest that money at that rate is not relevant for the discussion of IRR. It is relevant for the discussion around ROI. You obviously don’t double your money with ESPP.

But, it is a FANTASTIC IRR, and one people should take advantage of.

They shouldn’t be mutually exclusive. Because with ESPP you get all your money back after 6 months, you should be able to do both ESPP & 401K. Only the 401k locks your money up for a long time. It may take some cutbacks on expenses, but the right thing to do is to do both.

If I was forced to choose, I would say always do the 401K up to get the maximum company match. Very often, this is an immediate 100% return.

I would them max out ESPP, and after the 6 months, I would use the money to step up investment in my 401K and/or Roth IRA, depending on your situation.

Adam