Hacker News Reader

Comment by _sword | Hacker News Reader

Hi former Wall Street enterprise software analyst here. Financial models were, generally speaking, tightly managed by the companies being modeled. For each upcoming quarter and full year, management (typically investor relations, sometimes CFO) would hop on a call with my team and discuss where our estimates were vs. consensus. Sometime, "consensus" was an internal measure and not what was reported by e.g. Bloomberg or FactSet.

If estimates were particularly out of bounds from consensus, they would politely ask how we modeled their business, if we would like help modeling their company, if we had a particular reason for out of bounds estimates, etc. That was a firmly worded but polite way to describe that the estimates might need some review and adjustment.

View on HN

Hi former bulge bracket technologist/trader here as well. That seems like an extremely terrible idea due to the ability to pollute independent thinking by the model makers. What do you think about that?

I think that's the point though. Every growth software company wanted to have a beat and raise quarter, so they would manage expectations down to be able to offer exactly that, often no matter the quality of the earnings. As a sell-side analyst, if you wanted access to value-add with opportunities with corporates where you could get paid (e.g. non-deal roadshows), or even potentially banking business, you would generally need to have a good relationship with the company. The more divisive analysts would generally restrict contentious calls to one or two names that would generate call flow.

A lot of sell-side research work that analysts are paid for also focuses on information outside of estimates such as brokering investor sentiment or offering more details on channel checks in addition to what was published.

There was the whole era of “the whisper number” where analysts would publish one estimate and then feed journalists and certain customers another number. I specifically remember an analyst on CNBC literally say that he expected earnings on a company to come in ahead of his own estimate. Honestly, why is that number not your estimate then?

Whisper numbers are still totally around, but as buy-side expectations in my experience. I'd often field calls around earnings from investors who were trying to understand what everyone else was expecting from earnings results. There would then be further debates about what numbers would be good enough, or what long-only investors were expecting vs. hedge funds.

I remember (and will try to find) this one instance a few years back where one of the large banks issued a price upgrade shortly before earnings. This brought up the consensus prior to the earnings announcement causing the company to miss/underperform consensus which CNBC and other public news outlets weren't afraid use as the headline grabber...

The company management was pretty pissed, but it also shows how simple the public can be in interpreting the earnings miss/beat.

I no longer trade during earnings because it’s effectively random for John Q Public. A company beats across all estimates (eps, total rev, next quarter guidance) but drops 6% because a random non-public metric (like same store comp sales, or new subscribers added) that not a single analyst talked about that quarter was missed.

This is pretty common on the street. Not only for companies but buysides trying to manage estimates up or down.

Were parties aware that this was close to or a de facto reg fd violation (https://www.sec.gov/rules/final/33-7881.htm) and/or how was this handled internally?

Incidents and accidents notwithstanding, those involved in calls with and modelling for analysts and rating agencies are well aware of these rules. At least in my firm there is proper compliance coaching and continual awareness programmes about closed periods and internal continuity on what is disclosed publicly and when. Schedules are transparent and analyst calls in audio or transcripts are made available. Industry professionals even read these of competing firms, especially when there's trouble or a lol to be had. There's a lot that can go wrong, intentionally or not, but this part is quite a managed process.

Some companies like Microsoft were highly rigorous in managing information disclosure, and also in maintaining a lengthy quiet period. I could often hear investor relations at MSFT in the backgrounds of calls flipping through their internal playbooks of what they can and cannot say. Some other companies weren't as structured. I can't speak for how corporates managed disclosures internally, but I managed my own risk by typically publishing right after e.g. a corporate access event to reduce the risk of selective disclosure of material information.

I saw a handful of Reg FD filing updates and stock halts when new information was accidentally put out there by management. That was always embarrassing.

Could you explain more? That is like a hundred pages of regulations and I don't know anything about this.

Short answer: reg fd requires companies to disclose information material to investors to all investors. The previous post is suggesting that the close work with the bank analysts is conveying material information without proper disclosure. My hunch is that the legions is compliance lawyers at both the banks and at the companies have deemed this to be within the bounds of the regulation, but we’ll see if the SEC/US Attorneys agree.

Being cynical, I'd assume the "help" they offer is more likely to be stock market manipulation than disclosure of material information.

That's incorrect for these style of calls. The help is more akin to marketing - the company is arguing that they are doing better than the analyst thinks, and the analyst's company is offering their help to explain that to them and to other analysts.

That's exactly what I meant. They are not necessarily disclosing real information, they are trying to increase the stock price.

That isn't what stock market manipulation[1] means.

  Market manipulation may involve techniques including:
    Spreading false or misleading information about a company;
    Engaging in a series of transactions to make a security appear more actively traded; and
    Rigging quotes, prices, or trades to make it look like there is more or less demand for a security than is the case.

Presenting real information in a way that makes it more clear to investors is absolutely not market manipulation, and no interested party would ever claim otherwise.

[1] https://www.investor.gov/additional-resources/general-resour...

FWIW, my hunch is lots of information gets shared that the SEC doesn’t know about.

As long as breaking regulation is on average profitable the market dictates that regulation will be broken.

So depending on the risk/reward it most certainly can happen.

It makes me smile that you would ask this.

Where they aware everyone got bailed out in 2008, with nearly zero consequences?

The regulation you are expecting does not exist, in practice. Yes they wrote it down in a book. No, it's not real. It's as though I'm in a world of finance believers and I'm a finance atheist.

It does not exist.

Reply on news.ycombinator.com