Hacker News Reader

Comment by ThrustVectoring | Hacker News Reader

Back when I was fantasizing about what I'd do if I won the lottery, I looked into family offices a bit and concluded that there's basically no point as far as the investing advice goes. It's still likely a good idea for some of the ultra-wealthy for estate, tax, and philanthropic purposes, but on the investment side? The standard passive indexing approach used by middle class individuals scales in a cost-effective manner to the billions of dollars of assets. And as a bonus, you don't need to stake your fortune on a trust-based relationship. Instead, you get to base things off of institutions and audits - Vanguard isn't going to take the money you shoved into VTSAX and embezzle it through investing in their distant cousin's "company", even if it is a billion dollars.

View on HN

There is one point you haven't considered: Even though a pure indexing strategy is appropriate, it is not appropriate to use Vanguard for this. At the billion-dollar level, assuming that the index funds had an expense ratio of .05% (among the lowest out there) you would be paying annually:

1,000,000,000*.0005 = $500,000

For that level of expense, you could instead have a one-man office or other service provider that can buy the individual stocks comprising (or closely approximating) the index, and not have to pay any expenses other than his salary and trading costs. Meanwhile he can provide tax planning/philanthropic services as well.

VTSAX is 4 basis points of expense ratio. There are others with 3 bps and Fidelity has one with 0.

Even at 5 bps on a billion, I think you'd be extremely hard pressed to do everything Vanguard does for you for $500K/yr.

I'm at least a factor of 500 away from having to consider this question, but if you told me it would cost me $500K to have one fewer critically important person on my staff to deal with, that would be a good tradeoff in itself.

Yeah, I don't think it's that great of an idea. Any sort of operational or execution-related risk you introduce by replicating indices with single names could cost you a lot more than $500k and really isn't worth it.

There is additional value that can be added by forgoing an index fund, though. Tax management can be enhanced by rolling your own Total Stock Market index fund, since individual names can be harvested for capital losses, increasing the after-tax returns of your portfolio relative to a vanilla index fund. I'm not sure quite how to quantify that, but I'm sure someone has been able to do so.

Absolutely. There are a lot of reasons to run your own family office investment office.

"Because Vanguard charges too much" isn't one of them, IMO.

I thought you were only able to write-off $3k in capital losses per year against income?

If so, not worth dealing with for a billionaire, but definitely an opportunity for a middle-class robo-advisor.

You can write off capital losses up to the amount of capital gains plus an additional $3K. So, if you have $5MM in capital gains, especially short-term, it makes sense to sell off up to $5MM in capital losses.

VIIIX's[0] Institutional Plus Shares (with a $100 million requirement) has an expanse ratio of 0.02%, so that's $200k.

It would be hard to find someone at this salary level who can match the tracking performance of Vanguard. If they achieve a tracking error that's 1% higher than that of Vanguard's, that would mean an annual loss of approximately $800k compared to Vanguard (1,000,000,000 * 8% * 1%).

[0] https://institutional.vanguard.com/VGApp/iip/site/institutio...

This really is a no brained. With such a small fee with such a large investment you’d be extremely hard pressed to beat that performance (if all you wanted was to match an index).

Plus you can directly control the access to the account yourself. One fewer middleman to possibly scam you.

a one-man office or other service provider that can buy the individual stocks

In the given scenario, you aren't just hiring a guy to buy stocks for the Vanguard expense ratio. You're also buying all their financial and information security processes. You're buying risk mitigations like regular audits and SOX compliance and SP800-53 controls. To safeguard a billion dollars, that overhead is totally worth it.

The S&P 500 publishes changes to the index in advance of the changes taking place, so all of the affected stocks have prices that reflect the change at the time of the change.

Most tracking indexes, especially the ones Vanguard uses, do not do that and thus do not allow the markets to front-run them. If you adopt the strategy of "do what Vanguard does" and you do it immediately after Vanguard says they did something, you are already too late to get the prices that Vanguard got and can kiss at least .05% goodbye just based on that. I would expect to under-perform by at least .25%, if not more.

That's really funny. One man recreating Vanguard's business single highhandedly and doing triple duty with tax planning and philanthropic services and doing it for less money than Vanguard can do just the index funds. Thank goodness the knowledge to handle each and every one of those services, at an expert level, is interchangeable.

> For that level of expense, you could instead have a one-man office or other service provider that can buy the individual stocks comprising (or closely approximating) the index, and not have to pay any expenses other than his salary and trading costs. Meanwhile he can provide tax planning/philanthropic services as well.

But then you have to trust that guy not to fuck it up, while Vanguard has a stellar reputation and a whole office of people making sure things go as expected.

Tracking error against the index, auditing, security, and financial controls are more important than a couple basis points.

Honestly a much stronger argument is tax efficiency. If you're buying the individual stocks, then even when the index is flat you'll have some capital gains and losses. You can use both these to improve your tax efficiency - the losses can be harvested to offset realized gains elsewhere, while any appreciated shares can be donated to charity and repurchased with new money, effectively increasing your tax basis.

Still, you're likely better off just hiring people to do the tax planning, estate planning, and philanthropic services. It's essentially setting up a single family office with no control over your investments.

Agreed with what everyone else has responded to counter the claim that a one-man family office can reproduce Vanguard's VTSAX. Such a family office also has to contend with other giants trying to claim the G.O.A.T. title for asset management companies [1] with incredible scale efficiencies. The stakes are "for all the marbles of the game" high [2], and they're using all the benefits of capital scale they can scrounge.

The likelihood of a single person delivering index tracking performance better than these giants is pretty slim. Better to drop the tranche designated for broad market passive indexing into one of these giants' funds as an institutional holder, then have the one-man office project-manage leading education of younger generations, help the family leader clarify the ongoing family mission the office supports, the accountants' filing tax compliance at the individual and trust/foundation levels (likely in many different jurisdictions), grooming a successor, updating processes and procedures to improve auditability/accountability/anti-fraud detection, assisting with legal compliance, helping out with financing approved moonshots, etc.

[1] https://riabiz.com/a/2018/12/13/vanguards-asset-machine-wobb...

[2] https://www.bloomberg.com/markets/fixed-income

Don’t forget: keeping track of the dividends and cost bases of a hundred or thousand stocks will heavily impact your accounting/bookkeeping bill each year.

And once you talk about foreign stocks, you have different amounts of withholdings to account for, dividends that are not dividends (return of capital).

Indexing might make sense for a portion of your assets but if you have enough wealth to consider a family office, your investment goals are much different than a standard retirement account. Whether it's a different time horizon or simply having enough money to invest in assets that aren't available to you in your company's 401k, things are different.

What you are missing is that family offices usually farm out most of their vanilla public market securities portfolios to people like vanguard, but that 1/2 to 1/3 of the portfolio is private or alternative investments and that is where the investment team spends a great deal of time. Additionally, a lot of the "team" is often focused on things like reporting, tax, estate planning etc.

Yeah, what I'm saying is that you probably want to farm out 100% of the portfolio to someone like Vanguard.

> The standard passive indexing approach used by middle class individuals scales in a cost-effective manner to the billions of dollars of assets

Family offices are more like endowments and follow similar strategies that will have a portfolio that includes a mix of public equities, bonds, private equity, hedge funds, and real estate. Many of these investments are illiquid and have long holding periods, so aren't available to normal investors but can provide much better returns than index funds.

In addition, when you have >$100m in wealth, consistent, predictable returns become very important and index funds don't give you that. It took 8 years for the S&P 500 to recover from the dotcom crash and 6 years for the 2008 crash.

>Many of these investments are illiquid and have long holding periods, so aren't available to normal investors but can provide much better returns than index funds.

They can also provide worse returns. You don't know which. You can guess that they'll provide average returns, because the average investment gets average returns. After all, not everyone can be above average.

Passive indexing guarantees average returns. Before costs, that is. So not only is the indexing approach cheaper, but it's also safer.

>In addition, when you have >$100m in wealth, consistent, predictable returns become very important and index funds don't give you that.

Quite the opposite, IMO. If you do something straightforward like dump 100% of that into the S&P 500 and spend the dividends, you're looking at something like $1.8 million this year. Less in a downturn, of course, but that's still pretty difficult to spend.

And if you have known large expenses, fixed income is really efficient at guaranteeing you the money needed to meet those expenses.

> They can also provide worse returns. You don't know which.

You do. That's the whole point. If you invest in a basket of top tier VC and PE funds, they will destroy indexes over the next 10 years. Yale is a great example of this; over the last 20 years their average returns are 12% which is 50% more than what an index did (7-8%):

https://www.institutionalinvestor.com/article/b17qpx3nyyqywd...

>If you invest in a basket of top tier VC and PE funds, they will destroy indexes over the next 10 years.

I'd happily do a total-return swap of the S&P 500 against any basket of VC and PE funds you'd care to name that are currently accepting new investments. Assuming, of course, that there's some reasonable way of collateralizing and settling things at the scale I'm willing to risk ($10k notional), which I honestly doubt.

If beating the market was that easy, then everyone would do it. So either the top tier funds become closed to new investment after becoming top tier, or they stop over-performing for idiosyncratic reasons, or something. I'm using outside view logic here, I really don't care at all for the stories they tell. It's honestly downright dangerous to pay too much attention to the marketing material of investment funds - after all, Bernie Madoff consistently showed 11% annual gains. At the end of the day, the average investment must achieve average returns.

> If beating the market was that easy, then everyone would do it.

It's not easy. You need $100m+ of capital and a long time horizon to implement it. Very, very few people have that. It's less than 0.0004% of the population.

If I was a billionaire, I would be very very nervous just dumping my billions in a S&P 500 ETF. At that level of wealth, you really ought to have a portfolio manager who can slice and dice your exposure in advantageous ways.

I don't think active management makes sense for the majority of folks, but billionaires are exactly the kind of people that it does make sense for. With a billion or two you can probably get yourself into some pretty good closed-end funds (if you're savvy about it) that will probably deliver better risk adjusted returns than vanilla passive.

Personally, if I had the money, I would go with 2 Sigma or AQR. Both of their past returns streams are stellar and uncorrelated with the markets.

You also need to consider overlap of investments with the main source of income. For Bezos, it makes no sense to just buy a NASDAQ ETF since most of his wealth is already highly correlated with it. Either he hedges this effect out or skews investments otherwise. In either case he'll need knowledgeable people that can plan the effects on both financial and tax side.

That's also a likely reason why family office holdings often look very odd as they often only provide a hedge in case the main source of income dries up. And therefore, they cannot have any connection with it. Pure performance on a standalone basis is often only second priority.

> I would be very very nervous just dumping my billions in a S&P 500 ETF. At that level of wealth, you really ought to have a portfolio manager who can slice and dice your exposure in advantageous ways.

Why? A couple of billions should still be a tiny drop if compared to the market cap of S&P 500. What do those active portfolio managers provide to you?

You don’t consider 100% exposure to purely US equities a problem? It’s a huge risk, especially in this market.

With a couple billion you can get yourself some nice deals in a bunch of different asset classes. Also with that kind of money you can put a chunk of your money in illiquid investments that potentially could provide some great returns.

80% of companies in the S&P500 over time produce zero return. 20% of companies account for ALL of the returns. In the past these would've been driven by massive growth stocks like Amazon going 1000x over a couple of decades. Now, the main reason to IPO a tech stock seems to be because the company has reached its growth potential and you want to flick it off to derisk early investors, and often it plummets as reality sets in. Meanwhile, high growth stocks are captured in pre-IPO markets and the gains are all accrued to private investors.

I think the S&P500 is an entirely different beast to what it was 20 or even 10 years ago as a result of this.

If you were a family office your goal would be to get exposure to pre-IPO growth stage stuff as well as the public market.

It depends on your needs. You may recoup the perceived investment losses via tax savings or other factors.

Also, when you get into aging family members and generational wealth issues, you want to have controls to reduce your immediate control. Decades of accumulated returns can be wiped out by a bad decision.

If I had billions to worry about, I’d want to make sure I had assets that were more diversified than what Vanguard offers.

Err no it doesn't for large amounts not losing money is as important - see the previous comment about RCP.

You can also get into special sits and arbritrage ala Elliot and partners

Not refuting your assertion, but the concentration of power in tracking funds is a looming problem. The FT did a bit on it last week. Here's something similar from the motley fool:

https://www.fool.com/investing/2018/12/01/vanguards-founder-...

Only a looming governance problem, but not in any way a financial performance problem. Only the latter will trigger any real reaction from the markets, and Bogle thinks that won't likely happen until 70% of market under passive indexing at a minimum.

Governance concerns are not structural and addressable when it reaches a critical mass. I wouldn't be overly concerned, just put in a word when someone starts to lobby for regulation requiring delegating voting power back to index investors, and fund managers having to vote the preponderance of pro-rata voted shares.

Reply on news.ycombinator.com