The global chip shortage is starting to have consequences

No need to apologize at all. This is conversation. We all have much to learn.

Today, what you suggested is precisely what I do. I try to make sure there are at least three pin/function-compatible chips that can swap in for any given device. Preferably from different manufacturers. I also talk to distributors to get a sense of volume. I prefer to buy devices and components that are being manufactured and stocked in larger quantities. A silly example of this is that it is much easier to find a 47 uH inductor in stock than a 50 uH part. One has easy substitutes, the other can turn into a nightmare.

As for buying 10K microprocessors, again, that can be a tough decision to make. On the financial front, you could be talking about a $50K to $200K expenditure before you sell any product. In terms of logistics, if I have $200K in microprocessors in stock and I can't buy RS422 drivers I can't build a product. Which means that the decision of locking-up cash in the warehouse can quickly turn into a nearly all-or-nothing proposition. In other words, if you are going to stockpile microprocessors you might have to stockpile another $500K in parts in order to ensure that the investment isn't worthless if there's a shortage.

And then there's the issue of what you do with your nice pile of components if nobody is buying anything. As 2020 has proven, if you are in the wrong category, you could literally sit there for a year without selling much. That's what really hurts when you locked-up a pile of cash in the warehouse. We have a client who's business went down 80% last year. They had to shrink from 50 employees to three. They had to further shrink from a 100K square foot facility to a 22,000 sq ft building. And business is slowly crawling up. Had they made a huge cash investment early last year they would have been out of business by now.

In the electronics manufacturing business you have at least three tiers of manufacturers.

One is the super small shop that just sends everything out to contract manufacturers, along with parts they purchase themselves.

The next is the small-to-medium shop that graduated to having the CM provide parts. In other words, you design your product and fully trust your contract manufacturer to handle the supply chain. CM's will work with distributors to stock components and build boards. There is no way CM's are going to stock components clients don't need just to be sure they have a supply for a year's worth of boards. The only components CM's might stock in large quantities are parts others are using that are low cost. A simple example of this might be resistors.

The next level is a case where a manufacturer has enters into a contract with the distributor and the CM to have "bonded" inventory. They commit to buying a certain quantity of product --no matter what-- and, in exchange, the distributor and CM will inventory enough product to meet the demands of that contract. For example, you might commit to manufacturing 10K LED bulbs per month and need to ensure a supply of, say, half a million LEDs. You sign a contract and this happens. The advantage of this approach is that you are billed as product is delivered rather than for the entire half million LEDs you bonded. Of course, you are buying 10K bulbs per month. It's a machine, once it is set in motion you have to meet your obligation.

The next level might be manufacturers that do their own in-house assembly. I've lived in all of the above categories. The in-house assembly case can give you a lot of control and even lower your COGS, but you are now paying for everything pretty much upfront.

Once you start adding other component classes (mechanical, optical, etc.) things get even more complicated.

Each of these models has a financial formula associated with it. I have no idea where CCSI (the dog washing machine guys sit). My guess is it isn't a high volume business. I would further guess they make boards in batches of 100 or so (I could be wrong). When you don't know a pandemic is coming and the world is going to come to a halt, buying enough to make 100 boards a month is the right decision. If someone suggested they should buy enough to make boards for the entire year it would not sound like good advise unless the cost basis of those boards was such that it materially affected profitability in a significant way.

Business has become so competitive and fast that everyone pretty much ends-up adopting a JIT (Just in Time) manufacturing methodology. Anything else is suicide.

Here's another take: Do I invest money parking components in a warehouse for a year --just in case-- or do I put it into marketing, R&D and new product development? I think I can say that, under normal circumstances, it would be irresponsible (as learned the hard way) to park it in the warehouse. No crystal balls.

As someone else in this thread mentioned, I too wish there were more documented stories of business failures. That's where the real lessons for all of us lie.

In those days, dare devils didnt last long.

Now, they can crash and kill, and drive again in a few hours.

The tl;dr is that a shortage is when there are people who are willing to pay more than the market price for an item but something is preventing the market from raising prices.. either because of laws against price gouging, price caps, or because manufacturers have some other incentive not to raise prices.