Comment by MrWiffles | Hacker News Reader

Comment by MrWiffles | Hacker News Reader

I just tried to look at the tweet to see where this data's coming from, but "[...] quoth the server: '404'".

But really, forty percent!? For a statistical study that's an unusually convenient/round number, and for a state with the population of California, the chances of that statistic being accurate for any measurement methodology are astronomically small. Sure the dude could've been rounding it, which is totally fine of course, but I also question the validity of that statement because that's nearly half the entire small business segment for the most populous state in the nation!

And that just doesn't sound right, fundamentally. It's possible, sure, but c'mon, really? Near half?! I don't buy it, not without data from well-respected sources to back it up.

So I go to click the link, only to find it no longer exists. Which further casts doubt on this statement.

(18 hours after tweet linked on HN is when I checked Twitter; or in my world, "just now".)

One can't even judge if 40% is unusual. Most businesses fail. It could a healthy sign that a bunch of businesses were started. It could be a normal amount, or it could be really terrible.

Without more context, it isn't really informative.

Marc Andreessen spouting off at the mouth about things he doesn't understand?? I can't believe it! /sarcasm

Andreessen has never been associated with "spouting off at the mouth about things he doesn't understand".

What exactly does he not understand, that his betters do?

That 40% of California businesses failing is right on track with the normal failure rate of businesses pre-pandemic.

40% in 2 years is 20% a year. Considering the definition of small businesses, that’s relatively unsurprising. You’d actually imagine more than 1 in 5 giving up in a year.

In any case, 20% a year for first two years, compare SBA data from 2018:

> “According to the Small Business Administration (SBA) Office of Advocacy’s 2018 Frequently Asked Questions, roughly 80% of small businesses survive the first year.“

So, right on track?

> 40% in 2 years is 20% a year.

40% failing over 2 years is 22-23% failing per year—√(0.6) = 0.7745966...—but close enough.

This feels very wrong. ~40% of new small businesses will fail in 2 years. If 40% of all small businesses fail, that's completely different. If I model a small business surviving for 15 years on average, conditional on surviving the bumpy start, that would imply the equilibrium is

- small businesses started each year ~10% of all small businesses

- ~40% of those fail within the first two years; that's ~4% of the total

- looking at the more mature/stable small businesses, ~12% will churn over that same period

- total expected: 16% of all small businesses failing over a two year period

- therefore the pandemic represents a 250% jump

If you model a startup living for 10 years on average, conditional on surviving the first two years, the pandemic would cause a ~200% jump.

If you model a startup living for 25 years on average, conditional on surviving the first two years, the pandemic would cause a ~500% jump.

I saw that yesterday and it said the data was produced by Womply. I can't find the source though.

BUMP: could this be it?

https://tracktherecovery.org/

This is older news, but it notes that the permanent “closure rate” in California was 9.3/1,000 businesses. I’m guessing that’s a monthly rate but it’s not entirely clear to me. It also notes that LA had seen 15,000 businesses closing with half of those being permanent.

https://www.cnbc.com/2020/09/16/yelp-data-shows-60percent-of...

Probably more a sad state of professional contractoritis as small business

Reply on news.ycombinator.com