tanderson92

It isn't credible because the IRS IG has nothing to do with securities law and so they aren't the proper authorities. A simple read of their website shows it exclusively has to do with proper application of internal revenue code.

My comment didn't mention the publication venue. Failure to understand that the SEC deals with insider trading is fatal to any credibility the author or the alleged whistleblower might have had about such an accusation.

From TFA:

> “Back in 2003-5 a memo was created within the IRS noting who was permitted to participate in ‘insider trading’,” says this whistleblower. “The memo noted that all IRS employees in the executive branch and those one step below (territory managers, etc.) were permitted to participate.”

This is not who decides what is 'permitted', which is of course within the purview of the SEC. The piece states:

> Why didn’t I bring these allegations to the proper authorities to investigate? I did and was essentially told to get lost

No, the author went to the IRS Inspector General, which is very much not the SEC. The SEC has a very popular (oversubscribed) whistleblower program which awards millions to those who provide material information or tips relating to such cases.

This article is, ahem, not credible.

As a point of fact, the yield curve is not inverted (nor has it been for some time). Furthermore, forward PE ratios show returns far in excess of current government bond yields.

It's not knowable at this time why Musk made his comments. As a writer the best Levine can do is contextualize the situation and how it compares to norms of corporate finance, try to probe at possible plausible explanations for Musk's behavior, and evaluate the relative likelihood of each as well as the potential pitfalls. Of course none of the proposed explanations are really tenable. It isn't Levine's task to explain why Musk made the comments -- that falls rather squarely on Musk, or, perhaps, the SEC.

incognito doesn't allow cookies to be stored after exiting incognito, so in fact it does allow you to read as many articles as you wish simply by closing and re-opening the incognito tab. As another poster noted, FF incognito doesn't track cookies across tabs.

Over 25% of Vanguard AUM are not in indexed products.

You mean an answer you agree with. He did try to provide a correct comparison of the flows, contra to your initial post.

The author discusses the use of revenue as a proper company financial flow to use for comparison, provides an example of an article getting it right, and spends a significant amount of the article explaining the difference.

bogleheads.org -- DIY personal finance and investing discussion site.

Discussions can take you fairly far into the weeds but you learn a heck of a lot.

Another thing to note is that Fidelity and Schwab both have commission-free secondary-market treasury trading. This is much preferred to savings account rates. Oh and they allow you to participate in Treasury auctions too; you receive the so-called "high-rate" as a non-competitive bidder.

I see what you mean. The only thing I would add is that they do credit analysis on who they are lending to, and it's basically just institutional investors not fly-by-night short-sellers.

The way this shows up on your taxes is that Vanguard stock mutual funds (almost) never make capital gains distributions, while other mutual fund companies do. Capital gains distributions are taxed, and are inefficient as they cost you money.

It is true, but I think it's reasonable to suspect that principal-agent problems are less of a concern when you have non-collaborating individuals trying to extract money from an institution which doesn't structurally have a principal-agent problem than in the situation where the corporation itself poses a principal-agent problem with respect to its customers.

I already explained that while 60% of the revenue doesn't go to the fund investors, it doesn't go to Fidelity either. If you look at the Statement of Additional Informtaion for the fund, 56.7% goes back to the borrower (not Fidelity, the security borrower) in the form of a rebate. The profit motive doesn't explain this, at least not in the simple framing you've made of the issue.

It's also more complex than simply measuring how much of the securities lending revenue ends up back in the fund.

Most of that 60% for Fidelity's funds don't end up back in Fidelity's pockets either, but instead in the pockets of the people who borrowed stock from Fidelity investors in the form of rebates. You can understand this like: the borrower puts up 100-105% of the collateral and so should get some portion of the short-term rate on that cash. It's not clear why the disparity exists between Vanguard and Fidelity though.

iShares is truly different where they actually take 15-30% of profits for themselves.

Why do you say that index funds are supposed to be low risk? That's not really true; they in fact desire to have precisely the risk profile of whatever benchmark they target and seek minimal tracking error relative to that.

Anyway, loans are marked to market daily with collateral required. Details are here: https://personal.vanguard.com/pdf/ISGSL.pdf

It is absolutely a risk, but I think you err by saying index funds are supposed to be low risk.

If forced to use Fidelity as a brokerage I would prefer iShares; however, I would prefer to not use Fidelity. I do most of my investing in non-retirement accounts and as a result tax efficiency is a concern. For this, ETFs (by any provider) and Vanguard mutual funds are typically superior to straight index mutual funds. The reason Vanguard mutual funds are unique in this respect is that they hold a patent allowed assets to be comingled between their ETF and mutual fund share classes, so the superior tax efficiency aspects of the ETF can be shared with the mutual fund (and vice versa).

Also, Vanguard as a corporation is owned by its fund shareholders as opposed to external shareholders while Fidelity is owned by a private group. I don't trust any company enough to lock up my money with them and possibly have to pay capital gains tax to switch later, if they decide to raise fees or make the investment not align with shareholders in some respect. As whitepoplar says, they're a profit-seeking entity. Due to the inverted corporate ownership structure of Vanguard this is not a concern. This argument also applies to iShares.

> and Blackrock keeps ~half.

This is untrue. Per iShares(https://www.ishares.com/us/literature/brochure/securities-le...):

> Q: How much of the securities lending proceeds are paid to iShares ETFs?

> A:. iShares ETFs receive 71.5-80% of the income generated from securities lending depending on the asset class held. For example, iShares ETFs that predominantly hold US equities receive 71.5% of the income from securities lending. All other funds (i.e., fund-of-funds and funds that invest predominantly in fixed income or international equities) receive 80% of the income from securities lending. Once the aggregate securities lending income exceeds a certain income threshold amount, each iShares ETF will receive an increased percentage of the income above original levels, i.e. from 71.5% to 75% for US equity iShares and from 80% to 85% for all other iShares, for the remainder of that calendar year.

By the way, Fidelity isn't making money off of securities lending on these funds, or at least has pledged not to. I still wouldn't use the funds, though.

Here: http://www.investmentnews.com/article/20180801/FREE/18080999...

Fidelity has pledged that all securities revenue will return to the fund and not Fidelity itself, so no, that isn't the way they (claim, anyway) will make money.

> "Fidelity is not receiving any revenue from the Fidelity ZERO Index funds for securities lending. Nor is Fidelity currently receiving any portion of the income that is generated from securities lent out." --Fidelity August 1, 2018

However, their securities lending practice is significantly less efficient than Vanguard: their total stock market index fund returned only 40% of securities lending revenue to the fund while Vanguard returned 94%. The difference there is in broker rebates -- how much of the interest on loan collateral posted by the borrower goes back to the borrower. For some reason Vanguard seems to be doing a much better job on this (possibly Fidelity is much more aggressive in loaning out shares so receives less benefit).

> A good sparse matrix implementation is key to any serious numerics.

I agree with much of your comment, but there is a great deal of serious numerics which doesn't use sparse linear algebra at all. In fact there are entire subfields of numerical analysis which involve fast algorithms for dense linear algebra.

This is a commmon misconception: there is no $15k gift limit. There is a $15k annual exclusion of gifts that do not need to be reported. If you want to gift more than that, you simply file a form and the excess about the annual exclusion counts against your lifetime exclusion. The lifetime exclusion in 2017 was $5.49M (if you're married it effectively doubles). Oh and as of 2018 it is now $11.18M thanks to the tax "reform" bill.

The link was changed from the original Nature article which has the original title. The new link is unrelated.

You're correct that the innovation is really about sparse linear algebra (though not just solvers) done in hardware. However the advantages are exploited not just by way of solving linear systems for implicit time-stepping. They implemented a Jacobi iteration in hardware for solving a system arising from discretization of the Poisson equation, but they also used an efficient forward-map of the coefficient matrix to do explicit time-stepping for a wave solver. So they solve both a time-independent and time-dependent problem, but neither with implicit time-stepping (though there appears to be no technical limitation here, it's just what they focused on).

With respect (I didn't downvote you), there is a workaround: sci-hub (I checked that it is available there).

For others who want a comparison; the two papers are very different: the Nature paper uses entirely analog techniques to achieve high precision results while the newly-linked article is a mixed-precision approach. That is, the newly-linked article only uses analog to obtain a low-precision estimate which is fed back to a digital computing algorithm to produce the final high-precision result. The hope is that the overall time to solution is still faster because you do not need (hopefully eventually slower) digital computation to achieve the low-accuracy result. I am not an expert in this particular hardware field (I do applied math so I am used to solving the problems they do here in digital computers), but here is what the authors of the Nature article say:

> Recent theoretical and experimental studies have used memristor (including phase-change memory) arrays to generate an initial, low-precision guess (seed), and rely on an integrated high-precision digital solver to produce the required high-precision solutions from the seed solution. Such types of accelerators are certainly beneficial, as they reduce the number of iterations required by the digital solver. Determining whether memristor-based hardware can be used to directly perform high-precision computing tasks, however, will enable a better understanding of how broadly memristive hardware can be applied. Such knowledge will help pave the way to build more general memristor-based computing platforms, instead of special-purpose accelerators.

Right, the innovation is in using storage-compute collocation to allow common data operations to be structured into the way the storage arrays are set up. The paper is really about doing sparse linear algebra in hardware not PDEs (but of course after discretization that is all PDEs are--linear algebra).

Not a coincidence -- he links to the OP's link in yesterday's column when discussing precisely that similar problem. It is likely how the OP found this older article; I know I read it earlier today due to Matt's link.

It is an argument for private donor-based science instead of public investment.

This is an argument against public investment in science and a reliance on private donor-based science. Is that really the direction we want for our country and the entities in which we want to entrust the future R&D decisions of our country?

> it is trivially easy to beat this strategy by choosing an index that performs better historically.

Past performance is no guarantee...

This is a very subtle question which is far from resolved[1] in the academic literature; it is nowhere close to the slam-dunk you state it is.

[1]: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3122326