In the age of RSUs, graduating in the middle of a recession is financially incredible IF you managed to get the job before all the hiring freezes AND it is at one of the publicly traded big N companies with good trade volume. Your RSU price is lower, you get more stock units for the same dollar amount at the time of joining in your initial grant (arguably much more important than refreshers).
I have a friend who recently got his MSFT joining RSUs at $130 a few weeks ago.
Glad someone else noticed this. We are yet to have a crash from the coronavirus pandemic. You'd need another compound 50% off before I call it a crash.
Indeed, we're currently somewhat above the nadir in 2018 when we were all worried about the effects of 10-25% tariffs on $250 billion of US imports of Chinese products.
I don't know about the rest of you, but I am definitely not feeling as confident in the mid-term, or even long-term outlook as I was then in 2018.
I'm neutral on the market right now, I don't think it's entirely unfairly priced. NASDAQ is in the best shape of the indices because tech companies don't seem hugely affected by all this insanity. Most major investment banks are expecting a huge boom in Q4/Q1 and return to business as usual. COVID new cases are trending downwards across all major economies (except the US, to follow soon) [1]
I'm actually feeling better now than I was in October 2018. Rates are lower, tons of financial stimulus, a waning problem and a warming stock market.
Indeed but as virulent as it is, and with so many people having zero symptoms or mild flulike symptoms, we're all going to get it. The question is whether we get it first, or the vaccine is developed first. Either way, this will sort itself out shortly.
Both social distancing and herd immunity reduce the R factor. Thus nearly everyone may end up with the disease in NYC, but in less dense areas it can stamped out with vastly lower infection rates.
That alone could make the difference between 70% vs 30% of the US population getting even with minimal precautions.
I am hopeful we can suppress this disease before herd immunity develops in cities. But, once that happens they stop acting like dense areas in terms of disease spread.
It’s really a question of regional R factors under various levels of social distancing, population, and herd imminuty.
I have a pet theory that what we're essentially doing is running a huge experiment where we select for the least dangerous strains of the coronavirus. People who get a mild flu or no symptoms at all are less well quarantined and are significantly more likely to pass it on to other people. People who get the worst symptoms are tightly quarantined in hospital, reducing their chances of passing on the worst forms of the virus.
It's probably not true and it isn't really based in any biological knowledge but it kind of makes sense in relation to what I've read about how the Spanish Flu epidemic ended.
That's exactly what happens. Diseases don't "want" to actually kill their hosts; it kills them too. If evolution was selecting for more dangerous variants over time, we'd all be long dead; multicellular life might not even be possible. But it's actually selecting for things that live in the host indefinitely and spread over long periods of time, which leave the host alive in order to do so.
In the long term, the coronavirus will very likely either be entirely eradicated by human action or become just another cold... but first you have to survive to see that long term.
This is also why the worst diseases you see are zoonoses, diseases that spread from other species. They are co-evolved (if you want the term for this, look up "coevolution") with their hosts, then they jump into us and do the things that they "think" will just lightly affect their hosts and let them spread, and we keel over dead because those are the wrong things.
Apparently, due to having its own proofreading mechanism, the virus is very stable genetically. I heard that over all the different strains that have been sequenced worldwide, the maximum difference is just 15 base pairs!
The Spanish flu was selected for extra virulence, because it started during a war and the sickest soldiers were sent back from the front to hospitals. Usually when people get sick, they isolate themselves, so the weakest forms spread.
There's enough social distancing going on in NYC right now to prevent the majority of people from ever getting it. It doesn't seem likely that the social distancing will be let up enough (and subsequently not clamped back down) to the point of allowing most people to get it.
On the surface that seems to be true. NYC has ~8.4 million people minus those who left the city. ~1% of the population has already tested positive (81,803), and compared to yesterday (81,803 - 76,876) = 4,727 people are testing positive per day which would take years before everyone gets it. https://en.wikipedia.org/wiki/2020_coronavirus_pandemic_in_N...
More pessimistically, NYC’s CFR is almost 5% while in a growth phase when South Korea’s CFR is 2% after cases peaked when it was 0.6% in the growth phase. This suggests NYC is massively undercounting cases and the actual rate of infection could be something like 5-10x as high, with the virus reaching 50+% of the population in months at it’s current rate of spread.
PS: A more nuanced look taking into account differences in the overall and infected populations could narrow that estimate down. But, it should not be assumed South Korea discovered every case or that each country is counting deaths in the same way.
Death counts are a less unreliable indicator. Current death counts in NYC- over 7,000 using both the health department counts before Apr 5 plus the FDNY counts, tho even that still has to be an undercount- indicate that at minimum around the time of lockdown (Mar 22) at least 300,000 were "infected"- at a 2% CFR and MTTD of 15 days- but more likely closer to 700,000 or 800,000. That's 3 weeks ago.
With known new death counts currently in the mid to high hundreds, that adds tens of thousands of infections 2-3 weeks ago.
My take is- we are already today at over 1M infections in NYC. And we have a grim fork ahead- either we see a steep drop in death counts, or we see a steady or gradual decline.
A steep drop from lockdown means lockdown works, but also means we still have another few million in susceptible population as things start to open up.
If we only see a gradual or steady decline, COVID more quickly becomes the largest cause of death in NYC for 2020, and we know that lockdown doesnt work well enough to keep from progressing through the population, but we also know we have much less of a susceptible population remaining.
For sure we're undercounting. I had it and I'm not part of any official count because there's no way to get tested. My cohabiting girlfriend almost certainly had it, but she was entirely asymptomatic. Once you factor in all the asymptomatic people, in addition to all the symptomatic people who didn't end up hospitalized and thus likely couldn't get tested, the multiplier on the confirmed cases could easily be significantly more than 10X.
Pessimistically it seems that something like up to 10% of the city's population got it by the time that almost everything shut down, and now that almost everything is shut down, the spread has been vastly, vastly curtailed, as we see from new hospital admissions figures (which don't suffer from the same kinds of underreporting bias).
Well, if he had joined two weeks prior, his RSU price would've been $190, so he basically got 33% more RSUs. It's basically luck but I'm happy for him.
It's advisable to hold your RSUs for at least one year for a favourable tax treatment, so the 33% discount he got is still very good; he's likely not going to sell his RSUs in six months.
He got the offer and decided on his start date way back in August of last year so any luck he happens by is good luck. He could've chosen to be luckier if the stock price goes down further, but I'd say this is pretty lucky as it is already for something he can't control.
Edit: ignore the tax part, I think I'm wrong there.
I'm not sure what tax treatment you're talking about, but I'm guessing you mean converting short-term capital gains into long-term. RSUs are taxed as income based on the price on the vesting date, so if you sell when it vests there's no capital gains and thus no advantage in waiting.
That said, I think Microsoft still does a 1-year cliff, so they won't have anything to sell for a year anyway.
Yes. RSUs for publicly traded companies are for functionally equivalent to a cash for tax purposes. Not selling them immediately is the same as if your employer had given you a cash bonus and you bought shares of the company with the whole thing.
Anecdotally I'm still getting lots of recruiter spam from these types of companies on Linkedin and my current company that offers a good deal of RSU's is still hiring. It's not everywhere but in this particular recession and within those offers already offering lots of high value RSU's I suspect hiring is not slowing by a ton.
Recruitment spamming isn't useful for accessing hiring rate. Recruiters are always building their database of who is out there and gathering information.
I mean building your database via "I'd like to connect" message is one thing, getting specific job descriptions + being asked to start the recruitment process is a bit farther than that. I'm getting about the same ratio of type so far the past month. I'm not saying it's 1 to 1 but these companies are at least still actively hiring, and I have no reason to doubt that based on my company doing the same.
Gathering information is the recruitment process. It's just used to filter the good candidates out of bad candidates. The process is further used to get a valuation of what each candidate is willing or desiring to be paid and this is the main reason recruiters will always be working. I'm still being spammed as well and where the recruitment process is part of the spam. I work in academia which I enjoy and I'm not looking to leave my job but I do think the actual hiring will slow somewhat down. Too early to predict how much in all areas of engineering but I assume in a few months we will see.
I’m not sure how serious you are, but that gate-keeping formulation (that another community is “leaking”) is just about the most Reddit-esque thing you could say.
Are companies lowering the equity value of their offers currently? They might do that if they believe the stock will recover by 20-30% in 6 months, but I'm not on the job market so I don't know if companies are actually doing this.
I have a friend who recently got his MSFT joining RSUs at $130 a few weeks ago.