Trading bot that buys stocks bought by politicians is up 20% since May 2022

[Rebelgecko]

It's hard to find a calculator/data source that takes dividends into account, but it looks like if you bought an S&P500 index fund in May 2022 you'd be up around 18% depending on which day you purchased. So not very dramatic IMO

[paxys]

And if you bought it a few days later (on June 13, 2022) you would be up 20.5% today even without dividend reinvestment. Trying to derive meaning from stock returns over such short periods is meaningless due to the inherent volatility of the market.

[danuker]

Looks like VFINX returned 8.8% since 2022-05-01. How did you get to 18%?

https://totalrealreturns.com/n/VFINX,VBMFX,USDOLLAR?start=20...

[despideme]

The point is not if the politicians did better than they would have with another investing approach. It's a standard conflict of interest: the appearance of insider trading is corrosive because others see it and figure it's "how the game is played." So they need to do the same to keep up.

[signatureMove]

2% advantage is huge.

[dralley]

May 2022 was peak "we're headed for a recession" dooming

[its-summertime]

A buying strategy that can compete with index funds is pretty interesting, isn't it?

[ckardat123]

This is my tweet, thanks for sharing!

If anyone is interested, I built a dashboard that tracks the performance of individual congressional stock trades here:

https://www.quiverquant.com/congresstrading/

It also allows you to search for trades by stock, instead of needing to parse through thousands of disclosure forms.

I do want to mention one weakness of my data, which is that I don't currently parse hand-filed disclosures. Most politicians do electronic filings, which are easy to scrape, but some still file by hand. Working on a solution for that, which should hopefully be live before too long.

[sam345]

What is the lag between the trade and the public disclosure ? My guess pretty long so probably fairly useless data. [edit] The tweet's one example shows a month for that particular example "This came after purchases of up to $115K of $LMT by Representative Scott Franklin on September 12th, which were disclosed on October 11th."

So can't see how building a trading bot that trades on information at least a month old even if info at the time was good would give any big advantage. This strategy is going to be fairly useless imo over the long term even if you assume the politicians trading was better than the overall market.

[RandomWorker]

Pass it into chatgtp see what it comes up with.

[tootie]

You should track how their returns compare to indexes

[fsdadsfdsalkj]

Where do you get the data?

[pdq]

@dang, can you change the display for (threads.net) to be like twitter, where it shows the username to the right of the post title?

For example, threads.net/@quiverquantitative

[sp332]

Typing "@dang" does not actually do anything. Write to the mods using the Contact email at the bottom of the page.

[hn1986]

that's a great idea! did you message the mods?

[dmckeon]

Radical proposal: Rather than trying to police leakage of insider information, level the playing field for both politicians and fed employees by creating a dozen or so national mutual funds of at least 1,000 equity positions each that intend to model the US economy as a whole. Anyone can invest in them, but all US Federal employees and immediate family members are required to liquidate all other positions and hold only national funds. If the economy and funds do well, all federal folks profit. Distribute fund management among multiple fund management firms.

[atrettel]

I think that much of what you propose is already enacted in practice, at least for many federal employees who are not political appointees, etc. Elected representatives are another matter entirely.

Many federal employees do in fact have to limit what securities they can purchase or hold. For example, if I recall correctly, patent examiners cannot hold more than $20,000 in investments for an industry that they examine patents in, and they cannot hold more than $10,000 in stock for any company they may examine patents for. The guidance tells them to prefer index funds that capture the entire market rather than individual companies or sectors. This is explicitly to prevent conflicts of interest. If you are high up enough in these federal positions (without having to be a political appointee), you even have to complete a financial disclosure form (SF-714), and people do check these things for potential conflicts.

That said, many federal employees invest heavily in TSP funds, most of which are managed by BlackRock, so your last point about distributing them among multiple fund managers is not true in practice, since I imagine most federal employees have their investments heavily in funds managed by BlackRock.

[kbenson]

Wouldn't that just incentive those people to always favor the items in that fund, therefore providing an advantage to anything that makes in on the fund?

Given two competitors in a market, what if the existing but inefficient one is in the fund but the new one that is trying to shake thing up and/or do a better job is not? You might see a lot more protectionism, and not just against companies of other nations, but against our own new best companies.

[tzs]

What do you mean by "immediate family"? That usually means spouse, parents, children, and siblings. Sometimes also grandparents and grandchildren.

If you mean something like that then your proposal seems overly restrictive. I don't see any point in say the USGS not being able to hire a seismologist they want to hire because the seismologist can't convince his brother to sell his Microsoft stock and his adult daughter to get rid of her Apple stock. Nothing that seismologist would be able to working at USGS is going to influence those stocks, nor is his work going to provide him with any inside information.

[reidacdc]

Even more radical proposal: Make this restriction on senior corporate officers. Your salary can depend on how well the company does, but your investment portfolio depends on the broader market.

I mean, as long as we're using investment restrictions for social engineering.

[tootie]

It's called VTSAX

[ghayes]

There are significantly better options here, e.g. by moving assets into a blind trust while in office.

[conformist]

Unfortunately the stock market is not (reflective of) the economy? It seems reasonable to force them to diversify nonetheless.

[leoqa]

What about state and local gov employees?

[conformist]

The two graphs have visibly quite different volatilities. A number like 20% is not meaningful unless you also somehow state the risk, eg in terms of volatility or beta or other measures.

[tmpz22]

Taking the poster by their word they do a good job highlighting the delta between a politician's transaction and their disclosure of the transaction which seems to have about a month delay.

I personally really really struggle to trust the data represented in these tweets (err threads?) after the barrage of day-trade influencers we've seen in the last few years.

Would be interested to see a peer-reviewed version of this experiment...

[doitLP]

Is there a legitimate argument for keeping this legal and not pursuing an expansion of existing insider trading laws? The incentives here are horrifically backwards.

[Aunche]

I've yet to see any good evidence that Congress is outperforming the market due to insider trading. Even if they were, I don't think that the net loss of a few million dollars in market efficiency is such a terrible sin, especially when you consider how terrible they are at their jobs. I find it much more concerning that they're spending half their time fundraising rather than actually doing their jobs.

[ekianjo]

at the same time let's remove the immunity of politicians and prevent long terms that keep people in office until they die.

[the_gastropod]

The S&P 500 is up around 16% over the same period, without counting dividends. With dividends, you’d be at around 18%. So… this strategy isn’t particularly impressive unless it can continue this trend over several more years.

[satvikpendem]

One year is nothing, let's see how it does over 20 years and 50 years.

[kleene_op]

Maybe there is years of data available on that?

If so, someone should do an analysis of how much politicians pocketed since then.

[acheong08]

Did something similar for Malaysian stock market (bursa) and got ~26% since August. I think this is probably much more effective in corrupt countries.

(Our politicians and inside traders don’t disclose stuff obviously but there are signals that indicate an inside trade based on warrant stocks)

[rootsudo]

I’ve looked into this- basically when the congress member purchases or sell stock they have a month or so before they have to report it. It makes live trading impossible. But it is interesting to go back and see who knew what , when.

[dh5]

Separately, retail investors also outperform the S&P on average: https://www.wsj.com/finance/stocks/who-you-calling-dumb-mone...

> The average individual-investor stock portfolio has risen about 150% since the beginning of 2014, according to investment research firm Vanda Research, which began tracking the data nine years ago. That beats the S&P 500’s roughly 140% during the same period.

[strangesmells06]

How does one get this information about congressional trades?

[paulpauper]

not very useful. no comparison of S&P 500 and it stops in early 2023 and starts in 2020. a 3 year backtest that ends in early 2023 and ignores the bull market that began in 2023 is enough to prove anything.

[the_gastropod]

The post says “The market has been flat during the same time period.”, which is not true.

[phailhaus]

This is super interesting. The delay between the transaction and the disclosure may not be much of a handicap if lawmakers are making the purchases far in advance of the events to come. Has anyone run a historical backtest of this strategy?

[paulpauper]

It is hard to run a backtest when the actual strategy cannot be explicated precisely. For example, you would need to know the size of the account/capital. Just saying that a congress person sells $1 million of microsoft is useless unless you know when it was purchased how much capital total etc.

[sackfield]

I wonder how big the difference would be if the politicians close associates had to report their asset portfolios. I'm willing to bet there is a lot of "tips" being spread and favours repayed as a form of laundering.

[Animats]

This is running on info that's about 30 days behind politicians' trades. If it was running on, say, info that was only 24 hours behind, how would it have done?

[Imnimo]

If this isn't just luck, and it's based public and obvious information, why wouldn't the effect already be priced in by large firms trading on it?

[Joel_Mckay]

That is almost 3X to 4X times higher returns than most bonded investment managers.

Kind of sad really, but it is easy money for that 3% high-risk section of a portfolio =)

[olliej]

I still don’t understand why elected officials trading on non public information can’t be considered a intelligence breach

[missedthecue]

My unpopular opinion is that I think politicians trade on non-public info much less often the the average netizen thinks. There are high profile or notable cases, yes. But I'm talking in general. Pelosi for instance recently traded Roblox in size. What kind of info could she possibly have about that. But Twitter was up in arms about it.

The obvious thing to do from a policy perspective is simply require investing in index funds with sales scheduled well in advance. C-suites and bankers have to do it, why let politicians even give the appearance of corruption.