Hacker News Reader

Comment by arbuge | Hacker News Reader

"Local businesses recirculate dollars in their communities. An analysis of bookstores in Austin showed that, of every $100 spent in a locally-owned store (Book People), $45 were circulated back into the community, whereas only $13 made its way back into the community when $100 was spent at the nearby corporate chain store (Borders)."

Presumably the community local to wherever the Borders dollars end up is ok with this.

My point is that this kind of analysis stops half way. So what if business A doesn't channel its dollars to community A? Maybe it channels them to community B instead, or spreads them out over communities A, B, C, D etc. Perhaps there is a problem, but to figure out whether there is, the more important point to analyze is: when you add it all up and average over all businesses, is it a wash? After all, those dollars must be going somewhere, and I would think that they're making some communities somewhere pretty happy.

View on HN

This quote is one of those intuitively plausible things that turns out not to be true: "the more important point to analyze is: when you add it all up and average over all businesses, is it a wash? After all, those dollars must be going somewhere, and I would think that they're making some communities somewhere pretty happy."

It doesn't work that way. Think of the "lifespan" of a typical book-buying dollar:

(a) A dollar in circulation is created along side a corresponding dollar of debt. Banks are not getting dollars out of their vaults when they make loans. They are typing some numbers into a computer and creating entirely new dollars.

(b) The dollar passes from hand to hand in various transactions, eventually reaching our book-buying consumer.

(c) Someday, the dollar will likely cease to exist when it is used to retire debt. Even before then, the dollar can wind up "parked" (noncirculating) in large accounts that are held in reserve (e.g., in cash or in constantly rolled over treasuries).

A dollar spent at a big corporation like Borders is much more likely to reach state (c) sooner. This missing spending doesn't move to some other community, it just never takes place.

Also, if I buy a book from Amazon instead of the local bookstore, then I can spend the money I save (from Amazon's discount off the list price) on other local businesses, such as restaurants. And the UPS driver who delivers the book to my door presumably lives in the local area, so he's going to be spending much of his salary in the local economy as well. His local job might not exist at all if not for the extra demand that Amazon and other non-local companies make on delivery services. I'm skeptical whether the analysis that was cited accounts for all these factors.

Good point, but if you have centers of innovation and ownership, they will inevitably pull wealth out of the less successful places. This leads to inequality and a disharmonious nation-state (I sound like the Chinese government, don't I?). But really, there is a cost even to the "home of the borders" when their customer bases are slowly losing the ability to buy books.

Wal-Mart for example cannot survive without local economies providing people with enough to spend on their bread and butter. If we end up in a situation where most wal-mart towns are paid by welfare, then we're in a very inefficient system. It could get to be that way if everything gets centralized in commercial hubs.

I find the argument about fostering more investment in smaller companies compelling. There is a distributed systems aspect to it, which might enable higher overall returns because of increased network effect (e.g. a business not only benefits from a direct investment, but also from investment in other business in the same ecosystem)? Is there something here that bigger funds are missing, or is there data/evidence against it? Maybe it just becomes to complicated expensive to manage a ton of small investments vs a few large ones in a centralized manner? Maybe I'm missing something?

Reply on news.ycombinator.com