Hacker News new | past | comments | ask | show | jobs | submit login
Andreessen Horowitz Returns Slip, According to Internal Data (theinformation.com)
76 points by dawhizkid 9 months ago | hide | past | favorite | 40 comments

Be very careful. The IRR for a fund that is very young is deceiving. Fund V, for example, closed at the end of 2016. AT the time the documentation was put together, there could have been very little capital deployed and it doesn't represent subsequent rounds that have yet to be raised.

See link to a chart on returns by year/fund from the article: https://twitter.com/vcstarterkit/status/1173611439833006080

That looks pretty good to me. One dumb math question: when is a good time to compare against something like an S&P benchmark and should annualized rates or absolute growth be compared? In other words, is Fund I "cashed out" now, or could it's 44% return go up? Just doing simple math, adding 1 year to a CAGR calculation and doubling the FV of Fund I from 12B to 24B, seems to jump the return rate only to 48% from 44% (even though that more than doubles the profits). So would it be better to compare the absolute growth of the S&P to the growth of a fund, instead of the rate of return?

Venture funds typically take 10-15 years for the investments to cash out. It's been close to 10 years for fund II, but I believe they invested in Foursquare and AirBnB, so depending on how those companies do it could change their returns.

In general, it's hard to answer questions based on this chart. In addition to IRR, VC funds are judged on two other key numbers:

TVPI (Total Value to Paid In) which is the total value of all of the fund's investments divided by the capital LPs have put into the fund, both realized and unrealized. For the latter, LPs will often look at the investments themselves and make their own determination of value. Especially in this environment, many funds have investments in unicorns that are inflated and never end up being realized. A fund with a $800m TV may not actually look great if a large % of that is in WeWork and is based on a $40B valuation. This is also the case for IRR.

DPI (Distributions to Paid In) which is the total amount of cash a VC fund has sent to LPs divided by the amount of the LPs paid into the fund. At the end of the day, this is the most important number as it's what the fund's investors make, but it can take 10-15 years for a fund to completely distribute everything, so it's not that useful unless the fund has been around a long time.

Well, once cash is returned to the investor, using cash returned on the initial capital base to calculate a compound annual return is the best way. This would allow you to compare the fund’s return to a public market index over the time period. Before this happens, there is no best or one way.

I don't have access to the full article, but from the lede this looks like not very well researched journalism:

1. A16Z has 17 funds, with varying degrees of investing angles (early stage, crypto, bio, etc). So trying to do a fund by fund analysis is unfair.

2. IRR can be deceiving as it's time based. Some LPs invest based on "X Return" or IRR, and so cherry-picking one over the other is disingenuous without mentioning the other.

3. The larger the fund, the harder it is to have a higher IRR. A16Z keeps growing the size of it's funds (latest is $1b+). There are just simply not enough good deals out there to deploy that amount of capital. This is just like growing your top line revenue 50% from $1M to $1.5M, vs 10% from $10 to $11M. The former appears to have be semantically "better performing growth", when actually you made $500k more than you did previously.

4. The fact that they, a VC firm, are even returning their money means LPs will continue to invest. VC as an "asset class" is notoriously underperforming, with exception to the top 10% of the firms (which A16Z would likely be). Which begs the question, "so what?".

doing a fund by fund analysis is not unfair as funds are marketed to LPs separately. It does me no good that fund ABC had 30% IRR if fund XYZ that I am invested in from tbe same brand of manager is earning 1.5% IRR.

"IRR can be deceiving as it is time based". No, it's the other way. A return without mentioning how long it took to earn it is deceiving.

> A return without mentioning how long it took to earn it is deceiving.

Except that this is how some LPs want it to be reported, so how do you explain that? If you want to do a comparison to stock market returns ("seeking alpha") then sure, IRR makes sense.

I don't think this is poor journalism (I think The Information do a good job reporting industry specific information in a non click baity way). I see this as reporting on a player in an industry that a substantial number of their subscribers are either exposed to, or interested in.

But I agree with the rest of your statements. Better to have a VC investment that brings a 12% return or whatever than some negative returning European or Japanese government bonds.

> I see this as reporting on a player in an industry that a substantial number of their subscribers are either exposed to, or interested in.

Except the implication of the article is that A16Z funds are getting "worse" and thus underperforming ( = "A16Z must be a bad firm") making it newsworthy. I am far from an A16Z fanboy, but this is not newsworthy.

Normally business journalists refer to industry expert (for example, an LP analyst) to determine whether "is this bad or good?" rather than what appears to be "IRR is down, this must be bad".

Key points:

The funds the firm raised in 2010 and 2011 showed a net internal rate of return of 16% and 12%.

The results are a significant drop from the 44% return rate of its 2009 fund.

Did people think a 44% return was sustainable?

I would imagine that 44% return went a long way towards the vaunted status A16z had. In that, while perhaps not expecting to stay at such a high level of return, it's news because it wouldn't have been expected that such a drop (to the point where investing in common stock market index funds would have yielded a similar or better return) would have happened.

The point, as mentioned by the article, is to consistently beat out other similar investments at that scale (ie, index funds, private equity, etc). 44% isn't the competition, no. At a minimum 15-16% in the long run.

But otherwise at a more local level high risk capital expects high rewards.

And isn't 16% ans 12% still excellent? It seems better than almost anything else really...

Certainly not excellent - Obviously there is a lot that goes into it - but 20% would be closer to the average expected return on a VC fund like A16z based on the style of investing, not achieving that - or higher - would be a disappointing run.

This isn't an index fund we're talking about.

The S&P 500 has returned an average of 12.5% annually since 2011.

Would be interesting to know how much of an impact their bullish stance on cryptocurrencies and related projects had in this.

I've heard talk about this from VCs for years. Their reputation and results don't match.

A few thoughts spring to mind:

It seems like the more money you have, the harder it is to deploy it efficiently.

It would be interesting to know the influence of David Swensen (the CIO at Yale), who's put a large portion of their endowment into Private Equity and Venture Capital investment, and how other large funds might be mimicking his strategy.

People always compare VC vs the S&P500 but I wonder if there's a side benefit to VC in that it's not necessarily linked to stock market fluctuations.

Are VC returns in the aggregate going to turn to absolute crap over the next ten years as hundreds of new funds (with a new one popping up every day it seems) all grinding it out - or will we see the opposite, where a lot of these smaller funds have very successful first funds (partially constrained by the sizes they're initially able to raise), only to be dramatic underperformers as they raise second and third funds?

Finally, it seems like more money doesn't make for better results (past a point). The Vision Fund being example A.

With this and soft Bank going to great lengths to ensure WeWork IPO's, I think we will see more desperate moves by the investors in coming weeks.

Did you mean "ensure WeWork doesn't IPO"?

I think he means that Softbank can mark their investment at the last valuation if they stay private, but if they go public at $10 they mark a huge loss.

Is there a way to pay per article, or do I have to buy the $999 yearly subscription?

You can sign up for $1/m for 3 months if you have a referral link. I don’t have one handy, but searching on twitter you can usually find one.


Found in the Founder's twitter bio

What are returns of VC firms generally? Anyone has done analysis on this?

It is nearly impossible to do so given the private nature of these numbers but having worked in the industry extensively I can tell you the returns are generally atrocious and that no one invests in VCs to be fiscally responsible in the traditional sense.

Investors who put money in VC are generally so wealthy that by the time they are ready to invest in VC they have exhausted all other standard investment opportunities like stocks, private investments in mature companies, personal trusts and real estate and are simply looking for anything with a higher chance of return than a bond.

I’d estimate 90 out of 100 times the VC burns entirely through the money, 9 out of those 100 break even, and 1 out of 100 is profitable. 0.1 generate a return of something like Apple or Google.

That 0.1 in a 100 chance is good enough for the investors as they don’t feel any pain when it’s lost.

That’s why VCs have no interest in sustainable but mid-size businesses, only 100x opportunities. Their public marketing will push that it is because they are visionaries, the reality is that they have no interest or expertise in building mid-size businesses and the returns are so awful for their ‘visionary’ picks that without that 100x investment working out their funds would consistently be total losses.

Are you for real? This makes vc bros seem like the greatest con there is then. I'm sure the people running these vc's are making decent money from collecting fees on these random bets

Two Thoughts:

1. Consistent High Performers vs Everyone Else: As with colleges, grad schools, starting salaries, hedge funds, and a host of other things -- it would be good to have a top-20 "typical" return and an aggregate "typical" return.

2. Returns are only one aspect, returns correlation is another. Even post-fee returns at s&p500 rates would be awesome if they are uncorrelated to the rest of the portfolio.

Translation: Its set aside money for the purpose of gambling. If they roll a 7 then thats all the better, but its not critical cash flow.

Could you elaborate on what it means to have "exhausted all other standard investment opportunities like stocks, private investments in mature companies, personal trusts and real estate"? Is this due to some tax/estate laws?

Generally it's more about allocation and diversification. You can only put so much money in "standard investments".

When you already have a few hundred million in stocks, bonds, etc, the marginal benefit to putting another few million into the bond market is completely irrelevant--it's a rounding error in the overall portfolio. But putting those few million into a venture capital fund has the potential to generate a noticeable return.

It can also insulate you against structural shifts in markets. If WeWork were to fundamentally change the global real estate market, or some new battery startup fundamentally changes the energy landscape, investors in the incumbents can be left with significantly devalued portfolios.

Having a piece of anything/everything that might become the "next be thing" is a hedge against that.

It means they are already well diversified in all the traditional "safe" plays. "Exhausted" is maybe not the right word. Once you've got significant capital in the standard five ways then VC is your high-risk sixth play.

This is the best analysis I have ever seen on this subject. Kauffman is probably the only major investor that has enough exposure to do this (none of the Vcs they’re invested in are going to cut them).


This was posted earlier today https://news.ycombinator.com/item?id=20984000

The data is apparently a year old (see chart). A16Z has seen a number of large exits since then (Lyft, GitHub, etc)

paywall ha. Could we perhaps agree that whomever posts an article with a paywall should summarize the points made?

The paywall link should be disabled after 50 people unsuccessfully try to reach the site.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact