Arrrh, the mysterious 30% IRR.
If you’re a first-time entrepreneur making the rounds visiting VCs for the first time , you probably have heard about this phrase in a couple of occasions. More often than not you’ll be puzzled by its meaning. And even if you’re curious enough to have dug out the definition of IRR (Internal Rate of Return) by poking around the internet, you might still wonder why it seems to be always 30%. Exactly who dictates that? And more importantly, what does it mean to you as an entrepreneur?
Even the most casual investors know that rate of return is calculated by dividing a certain return number by a certain investment number. For example, in the case of a stock investment, the return comes in both dividends and capital gains when the shares are sold, while the investment is the amount of money one pays for the shares initially. A casual calculation takes the former number, divide it by the later and arrive at a percentage return rate.
However, this plain-vanilla return rate has many problems:
- It is not annualized — most people talk about a percentage rate of return thinking annualized numbers, so that they could be compared with the other interest-making vehicles such as saving account or government bonds (the semi-anual payment custom of US treasury makes this a bit murky though).
- It ignores the time value of money — the intermediate dividend or interest payments could be reinvested in other vehicle before one sells the securities and realizes a capital gain. These intermediate payments therefore are worth more dollar-for-dollar than the eventual capital gain.
Having realized the flaw of such casual calculation, it becomes easier to understand what IRR means.
IRR is an annualized rate (e.g. 30%) that would have discounted all payouts throughout the life of an investment (e.g. 16 months and 21 days) to a value that equals the initial investment amount.
Alternatively, if all the intermediate cash payouts could be reinvested at the same IRR (30%) and the position is exited in whole at the end of the investment horizon, then the casual calculation mentioned at the beginning of this article could yield a return number that, after being compound-annualized, would yield exactly the same annualized rate of return as the IRR (30%).
So why do VCs talk about IRR? Well, in fact not just VC, all the Private Equity people use IRR for calculating the investment performance. This has to do with the very peculiar cash flow patterns between the PE/VC funds and their investors.
Unlike liquid investment vehicles such as mutual funds or bonds, people who agree to invest in the PE/VC funds, called Limited Partners (LPs), sign contracts with the funds that would lock up the capital to more than 5 years.
Furthermore, the LPs won’t be able to put all the capital they commit to invest all at once into the funds at the beginning. Instead, they will have to wait until the fund managers, called General Partners (GPs), inform them to be able to wire money to the funds. Also, the amount of each wire is determined by the GPs and generally based on the need at the moment.
For example, a pension fund might sign up on January 1st, 2014 to invest in a PE fund, commiting to invest $20M, i.e. 20% of the fund. The fund eventually closed at $100M. The pension fund won’t be able to put their cash into the fund immediately, as would have been the case if they invest in mutual funds or other liquid vehicles.
On March 1st, 2014, maybe the GPs have finally reach agreement with a SME to buy it out for $45M. The GPs will issue a letter to each LP asking for an amount of the money pro-rata to the amount required to complete this transaction. In the case of this pension fund, that would be $45M * 20% = $9M. It will have to wire $9M to the designated bank account within a certain period, e.g. 2 weeks. The GPs then put all the money collected in this call to work by conducting the transaction. Over the life of the fund, the GPs won’t ask for more than $20M from this pension fund.
The call may happen multiple times, which results in multiple cash outflows from the LPs. Toward the end of the fund’s life, e.g 4~5 years later, the funds will start to return cash to the LPs either in the form of dividends or in the proceeds of a trade sales or an IPO exit. Since a fund typically invest in multiple companies, the cash inflows to the LPs, i.e. the returns, would also come in batches.
The time series of cash inflows and outflows over a fund’s life is then used to calculate the IRR, which is then used as a metric to evaluate the performance of the fund — the higher the IRR, the better the performance.
Technically speaking, there’s no such rules that it has to be 30%. Historically PE/VC industries yield an average of 20~30%, depending on the vintage years, i.e. the year when the funds are initiated. This range of return seems much higher than stocks and bonds. However, this is because PE/VC fund investments are much riskier than stocks and bonds.
In the case of buyout investments, the risk comes from leverage. In the case of startup investments, the mortality rate is just naturally very high, since startups sport unproven products and services, sometimes compounding that uncertainty with serving a market that has yet to exist or settle down in its value chain structure. And as even the most casual investors would know, the higher the risk, the higher the expected return should be to compensate for the risk. Therefore the 20~30% realized or expected returns of the PE/VS industries are not really that glamorous, especially if taken on a stand-alone (non-diversified) basis.
In additional to all this, investments in PE/VC funds also are illiquid and the distribution of the returns is historically highly skewed to left, with a fat tail on the extreme right. These all contributed to the historically high required rate of return, often stylized to be 30%.
Simple. Since the LPs come to PE and VC funds expecting to make a 30% return, the PE and VC funds will not make investments that yield an expected return lower than 30%.
Simple calculation would show that a 30% annual return would mean more than doubling the investment value in 3 years, which is generally the investment horizon for a series A or even seed round invesment.
Does this mean that VC funds will invest in a startup that projects to make a 2X cash-on-cash return in 3 years upon exit?
The answer is a resounding NO.
Because projected return is not equal to expected return. To see why, consider the following scenario analysis for a 3-year investment:
Note that the Base case is the unbiased projected return. Here I’m assuming a more generous 75% fail rate described by Wall Street Journal. I also keep a 1 cent return for every dollar invested in the Down case, a.k.a. fail case, only so that IRR calculation is meaningful. In reality the recovered rate for equity investors is 0%.
What is the expected value of this investment then, given an unbiased projection of 26% in the Base case?
- Expected cash payout = 1% * 10x + 24% * 2x + 75% * 0.01x = 0.5875
- Expected IRR = 0.5875^(1/3) — 1 = –16.2%
So instead of generating the positive 26% IRR in the projection, the expected IRR is –16.2%.
The key here is with startups, the probability of achieving that base-line projection is much lower than compared to a mature company in a mature industry — the upside even lower. While a mature firm faces mostly operational risk and firm-specific risk, a startup faces not only the two aforementioned risks — AND all the others.
This means that if a startup projects to make a 2X return in 3 years, most likely no VC will invest in it, because more often than not that 2X return won’t happen due to all the risks. And if the VC invests in 20 such investments, the overall return will definitely be lower than 2X in 3 years, i.e. lower than the 30% required rate of return.
So what do VC funds invest in?
Ideally, it should be the companies that project a 10X or 20X return for this round of equity investment. Consider the updated scenario analysis below:
- Expected cash payout = 1% * 20x + 24% * 10x + 75% * 0.01x = 2.6075
- Expected IRR = 2.6075^(1/3) — 1 = 37.6%
As can be seen, the startup still has 75% chance to fail completely and wipe out all equity values, but since base-line return now is 10x, the expected return is revised up to a respectable 37.6%. That would then be a viable investment for a VC fund. Further conversations would then be worth the investment manager’s precious time.
You really have to ask yourself a question: is your startup really a venture type of startup?
I mentioned that the 30% IRR is demanded because the startups face all kinds of risk. If your startup indeed faces all kinds of risks and you could only project a 30% annualized return — which as we demonstrated earlier does not lead to a 30% expected IRR — then you’re in the wrong business. It’s not worth doing it even with your own money. You might as well put that money into the US government bonds.
For example, maybe your startup is working on a connected object product that, according to your vision, will revolutionize the way people do their morning jogging. Well, revolution is risky — as most unfortunate French who lost their heads at the guillotine post 1789 could testify. People have been jogging for thousands of years. You have to hit all the right buttons to induce them to change their jogging behavior and you also have to find a way to profit ENOUGH from this change. Failing that you might run out of cash and go bankrupt. In this case, if your projection of return is much lower than 10X or 20X, it’s better you go get a job and use your own cash to buy some government bonds — in fact, even the Greek government bonds should look attractive compared to your startup.
On the other hand, if your startup actually doesn’t face that much risk after your analysis, that means the 30% projected return might be good enough to compensate for the risk. However, in this case you are not the target of VC investments — there’s a reason why the word “venture” is in the term “VC”, despite whatever is implied by the Armani suits worn by the managing partners to the IPO ceremony of the homerun portoflio companies.
In French, « Venture Capital » is translated into « Capital risque » — I believe you don’t even have to google this French term to know what it means.
So if your business is actually NOT a venture business, you should not waste your time trying to convince VCs that you will yield 30% IRR for them. It’s better to go to your in-laws with hat in hand asking for their sponsorship, assuming you’re still in good terms with your wife 7 years after the wedding ceremony.
But be careful there: sometimes the required rate of return from you father-in-law is even higher than the 30% IRR demanded by a VC.
Worse, usually cash settlement is not accepted.
Originally published at http://www.jmyang.com on June 26, 2014.
