Comment by vladd | Hacker News Reader

Comment by vladd | Hacker News Reader

Warren included several gems in this year's annual letter. Below, a small preview:

- discussion about December's tax change (>> The $65 billion gain is nonetheless real – rest assured of that. But only $36 billion came from Berkshire’s operations. The remaining $29 billion was delivered to us in December when Congress rewrote the U.S. Tax Code <<)

- the new GAAP accounting standard that will produce huge quarterly swings in Berkshire reporting in the quarters to come.

- 2017's frenzy in high purchase prices for American enterprises, and the side-effects of using debt to finance them. (>> If Wall Street analysts or board members urge that brand of CEO to consider possible acquisitions, it’s a bit like telling your ripening teenager to be sure to have a normal sex life. <<)

- payments made by Berkshire for hurricane insurance.

- strong discouragement to borrow in order to buy stocks (>> the strongest argument I can muster against ever using borrowed money to own stocks. There is simply no telling how far stocks can fall in a short period. Even if your borrowings are small and your positions aren’t immediately threatened by the plunging market, your mind may well become rattled by scary headlines and breathless commentary. And an unsettled mind will not make good decisions. <<)

- details about Warren's bet against hedge funds when compared to S&P indexing.

- risks in owning bonds versus stocks.

Overall, I found the letter to be a useful reading for those passionate about investing or financial self-sustainability.

"- risks in owning bonds versus stocks." This article summarizes this point nicely, https://www.marketwatch.com/story/warren-buffett-says-famous...

People are not wired to properly consider long-term risks. The risks in the ancestral environment were entirely short term - literally, the tiger in the bush. At worst it was a drought or poor season, with a series of short-term food security crises.

When people invest, they are worried about things like a market correction wiping out half their purchasing power should they decide to sell and spend it next year. These events are protected by owning bonds and cash. But the real insidious killer is the long-term risk of sitting in the sidelines as bull markets go by and wipe out your ability to purchase long-delayed spending on the cheap. Or in other words, the risk that today's price is the lowest you'll ever be able to find the S&P 500.

I'm worried about that last scenario enough that I'm considering changing my investment strategy from 100% global stocks to include buying some long-dated S&P 500 call options. If the market goes down my future cash flows buy my retirement spending on the cheap, so losing a relatively small amount of money on call options that expire worthless is fine. If the market goes on a tear and stays up, though, I'm in a pretty rough spot.

> People are not wired to properly consider long-term risks. The risks in the ancestral environment were entirely short term - literally, the tiger in the bush. At worst it was a drought or poor season, with a series of short-term food security crises.

This kind of pop evo psyche just so story is not any different than stories about how natural phenomena are the result of gods playing capricious games with humanity. It comes from the same place and has the same validity.

Your criticism seems irrational to me. Gods playing capricious games is not a plausible explanation for anything. Human behaviour reflecting ancestral selection pressures is at the very least plausible.

I think it's actually just as irrational to try and explain a series of complex behavior with one central premise. The truth is humans are complex. All that "pop evo psyche" stuff is just a story. It makes sense, it's vaguely plausible, but that doesn't actually make it true unless you only look for examples that confirm the story and ignore all the ones that don't.

You think it's just as irrational (as ascribing supernatural causes) to build a scientific theory? Who is denying humans are complex? Who is trying to explain everything with one central premise? The truth or otherwise of the "story" is a matter of debate. You seem to be wedded to the idea that the story is definitely untrue.

I don't see any scientific theory here. How exactly would we go about testing the relationship between risks in the ancestral environment and not being "wired properly [to] consider long-term risk"?

I'm not sure, but presumably all branches of science that deal with conditions in pre-history have testability challenges. Doesn't mean you can't build theories, reason about them, devise ingenious and indirect ways to test and experiment. Humour me. I assume you don't dispute that modern civilization has been conincident with only a tiny fraction of human evolutionary history. Why isn't it reasonable to assume we are well adapted to pre-civilization environments and that sometimes this is not helpful in modern life?

> Humour me.

It's exactly my whole point that it is unreasonable for me to do so. ThrustVectoring was trying to appropriate the good will and prestige that science has built up without doing the hard work that's necessary for his argument to actually deserve that good will and prestige. It's a kind of free riding.

I think that's an extraordinary stretch - but okay, if that's your point that's your point.

Edit (added): Sorry that's probably a little uncharitable of me. We've rather been talking past each other. From my perspective I've felt I wanted to discuss the science and you've been saying no, we can't discuss it, it's not science. But that's just my perspective.

I don't get it, if you're 100% on stocks, won't your portfolio rise with the market? Or are you worried specifically about your (lack of) exposure to the US market?

My investment account balance is low compared to my lifetime future earnings - even at 100% stocks, I'm arguably under-exposed to equity bull markets.

I don't follow the logic. What does your future lifetime earnings have to do with the returns of the market?

Sounds like the fascinating but terrifying 'mortgage your retirement' theory: https://www.bogleheads.org/forum/viewtopic.php?t=5934

Thanks for that! I have no idea what's the lesson to take away, but it sure was an interesting read.

If a bundle of stocks worth $1MM at retirement is worth $100k today, then in order to save up $1MM at retirement I need to earn $100k more than I spend. If the stock market goes up 20%, then I now need to earn $120k more than I spend, putting retirement further away. Call options protect against this, giving relatively cheap insurance against huge market upswings.

I enjoyed that article, thanks.

I want 1 year of expenses in cash, 2-4 years of expenses in bonds and the rest in stocks. Then I will have the courage to wait out (or buy into) the inevitable stock market crashes. This is my ideal retirement holding--while I have labor income is a different story.

I don't buy bonds for the return, I buy them for safety of principal. So, the risk of lower return is absolutely acceptable to me.

PS Please don't take investment advice from random people on the internet, definitely including me.

I have similar plans oddly enough.

2 years in Series I bonds, 1 year in cash, the rest goes into a very aggressive investment portfolio of equities and real estate.

It's not odd, from what I can tell that's the current standard investment advise: safety cushion in cash, then a mix of stocks and bonds, with the proportion of stocks being inversely proportional to one's age.

You are only allowed to buy 10k in I-bonds per year and you will forfeit part of the interests if you hold them for less than 5 years.

I am aware. I have been buying them ever since I ran out of tax deferred retirement space. It's as much for expanding my effective tax deferred space as it is a secondary cash reserve.

Secondary cash reserves all have similar problems.

How are you going to buy stocks when the market crashes if you are already invested as much as you can?

Edit: nevermind, I guess you mean with the incremental savings.

Yes, or possibly shifting some cash or bonds to stocks. But definitely letting existing stocks ride out the downturn.

People often write this as if investing in a "crashed" market is something trivial to spot. Sure you might be able to see prices today are lower than a week ago, but how do you know they aren't going to continue decreasing yet another month?

Dollar cost averaging is a decent strategy in a bear market. You basically buy more and more shares for the same money as it goes down. You’ll have bought more shares at the bottom them at the top.

In other words, set up buying on a schedule and stick to that schedule, purchasing the same dollar amount each month. Don't deviate from your schedule because you cannot time the market.

Why does it matter if they are decreasing next month? The goal is not to buy at the bottom of the market (if I knew how to spot market bottoms, I'd be on my private Caribbean island), but rather to keep buying into the productivity of the entire economy.

A bond crash can cause capital loss

Credit blowups can cause capital loss, risk-free bonds can never have capital loss if you hold to maturity.

Unless you take inflation into account.

My personal theory is that there are currently a lot of people waiting for the stock market to crash. Which is why it might not going to crash anytime soon (see the dips beginning of the year).

>> I don't buy bonds for the return, I buy them for safety of principal.

Stocks and bonds are not as anti-correlated as you might hope. The main benefit is diversification, and in that regard, it's really just lightening up your load of stocks - not actually the introduction of bonds. You may want to consider additional instruments for safety purposes.

Thanks for the suggestion. What other instruments would you suggest I research?

Edit: I should note that when I am saying bonds I really mean the type of bonds held in this mutual fund (for example): https://personal.vanguard.com/us/funds/snapshot?FundId=0132&...

This makes me invest in crypto even harder. It is the natural extension of this rule. Crypto is more risky than stocks but the greater yields vs. stocks more than make up for it. You can say "you could lose it all in crypto" but that's what the bond people say about stocks and the math is showing that to be completely backwards. Opportunity cost is the real killer.

greater yields

Are they? What's the average yield across the whole cryptocurrency market vs the US stock market?

It's very odd to hear the term "yield" be used to describe capital gains. Typically this term means return on investment from either interest or dividend income. Ie, the "safe" part of the return. For example it would be very strange to say an investment had negative yield (though it might have negative return), since yield is the money the investment "throws off" to the investor. If you are never planning to liquidate the investment, yield (and potential changes to it) is your primary concern.

The only valid form of yield in crypto I think is earnings due to Proof of Stake deposits.

>> greater yields

Past performance is no indicator of future returns. I am long cryptocurrency as well but the main attraction from an investment standpoint into cryptocurrency is twofold:

1) Volatility is incredibly high (useful for various reasons; does not come without a downside, also obviously)

2) Yield curves of US equities are trash

A portfolio that is predominantly cryptocurrency is... pretty silly. It would be hard to reasonably argue against a majority share of equities and basic portfolio advice from someone like Buffett, with heavier weighting on higher variance items like cryptocurrency.

I've been punting crypto a little, bearing in mind Buffett/Grahams Mr Market story where "... the more manic-depressive his behavior, the better for you." It's hard to find more nutty market behavior than with crypto. I daresay Buffett is bearish on crypto and has said he would buy 5 year puts on the whole lot if he could.

(http://www.alphavest.com/an-essay-on-ben-grahams-mr-market-b...)

Quote: ... A poll of investment experts asked in late 2007 to forecast long-term common stock-market returns likely would have seen them guess, on average, close to the 8.5% actually delivered by the S&P 500,” he said

I fear people may misinterpret what Buffett said esp. for those who don't have memory far back enough. In 2007 I recall one could buy long term CDs at 7%. These were basically relatively riskless guaranteed returns for small investors. Nowadays the comparable returns are a little above 3%.

During the financial crisis many credits of financial companies were selling for a fraction of their face value. For example AIG bonds, or Bank of America subordinated debt. You could buy them for a whopping nominal interest rate north of 15%.

Even as late as 2011, during the muni market scare, one could buy California state taxable bonds and and California school bonds (federal and state tax-exempt) at about 8% interest rate. If one were a little more adventurous (and read up on the laws a little) one could have bought California muni bonds backed by redevelopment tax increments at close to 10% (and insured as well).

All these were safer than stocks at the time and delivered comparable returns. They were not meaningful if you have a portfolio the size of Buffett's, but they were good opportunities for small investors.

Actually Buffett did a very profitable deal with Bank of America in 2011:

https://www.bloomberg.com/news/articles/2017-06-30/buffett-t...

Right. There are opportunities in equities not available to small investors and there are opportunities in bonds not available to whale-sized investors.

Where would you screen for the bonds you mentioned in this thread?

E*Trade offers access to BondDesk. Vanguard has online bond buying access as well.

Thanks, will check out if Fidelity has something similar.

I remember Fidelity had online bonds buying as well. As Buffett said, today's bonds are fairly expensive, offering little upside. I just think people should also consider the implication for equities.

That’s why it’s always good to have cash. I still have some NY GO and PIT bonds paying a little over 6% tax free from that era.

Making money on AIG and Bank of America bonds really required intelligence to understand why AIG of BoA wasn’t going to be another Lehman.

Intelligence yes, but the risk was also commensurate with the returns - or more specifically, the volatility of what the end result would be.

It's not at all clear what the risk-reward were, without in-depth information. After all everybody except Warren Buffet had backed away at that point.

Right, I agree. Hence, the volatility.

The bonds were significantly more certain than the commons, since owners were generally not subject to dilution risk. I was not referring to the height of immediate post Lehman uncertainty. Even after the Treasury Department bought preferred shares from the financial companies, thus subordinating the US Government to the bond holders, some of the bonds were still trading at very significant discounts (> 50%) to face value. I am not advising anyone to take risks in such investments. However if one can't deal with credit risk/reward at such advantageous terms one also shouldn't fancy that one can deal with the stomach-churning volatility of equities at such trying times.

If you're comparing the bond and the equity tranches of the same single company then I agree. Most people's alternative opportunity however is in, say, the total market on the equities side. It's arguably easier to reason about the long-term future value of the entire US equities market vs. whether a certain haircut on a single company's bond is sufficient compensation for the credit risk, especially when it depends so much on arbitrary and unprecedented political decisions.

This is why information is money. Warren Buffet may have known or understood the implicit political assurances of which banks were "too big to fail." Under such circumstances any idiot can pull the trigger.

While I tend to agree that equity basket is generally better for the uninformed, one should realize that this is also dependent on the faith in the politics. Generally it is difficult today to find a company that is not leveraged outside of tech and tech companies are leveraged in their own ways. However if one takes the attitude that politicians will take care of things, just beware that such faith will be sorely challenged at trying times.

>> Nowadays the comparable returns are a little above 3%.

Well below this for smaller investors. Yield curves are getting hammered. Generally about 2.5%, and that's through smaller banks.

> The ample availability of extraordinarily cheap debt in 2017 further fueled purchase activity. After all, even a high-priced deal will usually boost per-share earnings if it is debt-financed.

There are lot of risky bets being made on back of cheap debt. I don't know how long will this sustain - Tesla, NXP, Broadcom and Qualcom to name a few - It will be interesting to see how things turn once the rates start to turn.

Here's a good podcast episode by the person who took the other side of that bet: http://investorfieldguide.com/bet/

I’ve been torn about his comments of not using leverage to buy stocks. Berkshire uses insurance float which has risk. I guess they feel that this risk is small and supported by the 400 billion reserved to pay out.

Is attaining debt at a 4% rate mortgage riskier than utilizing their float?

A 30-year amortized non-callable loan is much less risky than insurance float. It's also much less risky than a margin loan (which is at a correspondingly lower interest rate - best margin loan rates for small balances is roughly 2.6% at InteractiveBrokers, last I checked).

"Don't use leverage" is good advice for people who do not have the skills and experience to properly manage risk (which is to say, over 90% of the population). The proper answer is something like a long explanation of the Kelly Criterion, risk of ruin, how to evaluate the very fuzzy notion of your own career's job security and social safety net, and a ton of other factors.

Circling back to your question about a mortgage, though - just don't become "house poor". You want to have free cash flow above monthly expenses. Outside of that, on a 30-year fixed-rate mortgage basis, feel free to borrow and invest literally every dollar the bank approves you for. Especially if the loan is single-action or your investments get funneled into creditor-protected retirement accounts.

> Especially if the loan is single-action or your investments get funneled into creditor-protected retirement accounts.

That is a very interesting strategy. Do you have any recommendations for books or other resources that discuss this?

I don't really, it's just a matter of risk avoidance, paying attention to what happened post-2008 housing crisis, and understanding the consequences of strategic default.

If you don't pay your mortgage, it's generally either impossible or not worth suing you for the outstanding balance. Usually what happens is the bank spends a few months going through the foreclosure process, trashes your credit score, takes your house, and evicts you. This is bad, but if you owe $400k on a house that's worth $300k, it's less bad than continuing to pay your mortgage and dump an extra $100k down the drain. You save up first + last + security + moving costs for an apartment or renting a house out of the freed up cash flow, deal with the fact that getting credit is going to be difficult for a while, and move on with your life.

This description of (individual / personal) "strategic default" is interesting, maybe because it strikes me as a theoretical approach that is extremely rarely employed in real life. Corporations can and do consider bankruptcy a rational option under various circumstances, but -- maybe for a mix of cultural and legal reasons -- for individuals, there's a stigma, a sense of shame and failure, and it's viewed as a desperate measure of last resort.

I know, citation needed...

Using the float shares the solvency risk with borrowing, but they aren't paying 4% on the money.

Correct, they are paying essentially 0% interest in exchange for the policy risk.

My interest is that the policy risk according to Berkshire actions is less than 4% debt they could likely attain (or better). In other words - they are recommending investors not to use leverage, but in a sense they do use it against policy risk.

The trade off between debt and policy risk is interesting as both greatly increase returns by leverage.

The risk is not on the debt you take, but on the stocks you buy. If the stock market halves next month (it could happen), will you be ok?

Edit: I guess a secondary concern is the risk of interest rates going up, will you be ok if the cost of servicing the debt doubles?

>> Edit: I guess a secondary concern is the risk of interest rates going up, will you be ok if the cost of servicing the debt doubles?

Are people seriously considering borrowing against adjustable-rate instruments to buy equities?

His historical leverage has been 1.6:

http://docs.lhpedersen.com/BuffettsAlpha.pdf

Best take away (though not expressly stated), one never needs more than 20% annual return if compounded for 50 years! $19 to 211,000 ! That’s more than the networth of 50 % of Americans. Compounding, Still the greatest lesson in the world.

>> one never needs more than 20% annual return

Talk about the understatement of the century. 20% annual return, that's it? Should be easy...

Reply on news.ycombinator.com