Recently I had one of those moments where something you have lived with about for >10 years all realigns and you see it differently. In all my previous labs, incl undergrad, my PI's always seemed to be near the edge of running out of all their money. Grants were obtained in the nick of time, things looked bad and then got worse, post-docs were hired despite the dire straits, etc. Then I started to manage a lab, and started to have people ask me for things that cost money, and I realized the value of telling everyone how close to the brink the lab is. That there is a never-ending strong pressure to tell the underlings that the money situation is parlous. I'm not saying it's a lie. Things ARE pretty much always dire, if you look at it from the right pessimistic perspective. And, running a lab, there is an incentive to take that perspective.
Telling people you are close to broke makes them less likely to ask for things, makes them likelier to think first before asking for inessential luxuries, puts them in the frugal frame of mind. It provides an incentive for people to apply for F and K and NSF awards, and to not just phone it in. It gives you a no-hurt-feelings excuse to reject lab applicants at all levels, and even provides some help in ushering out bad students + post-docs if it comes to that. It's a flexible control system: if you have to spend, you can just say, "oh we found a small pile of money", or you can blame the department accountants for misleading you, and then say, "but now we really are broke."
This deceit is helped along by the fact that it's really hard to know exactly how much money one has. Grants are unpredictable, money is budgeted into different accounts with weird rules, and money gets allocated without appearing on the budget until later, etc. As far as I can tell, it's basically impossible for lab members to guess how much money there is. I had no idea what a start-up was when I was in grad school, and it was embarrassingly late when I realized that an R01 is 249 per year not for a lump sum to be spread over 5 years. I also wasn't sure whether overhead came out of that 249 or not. And so on. It wasn't too relevant to me at the time - back then I was focused on learning the science. Now, I worry about the money. And I try to spread that worry around, in overly pessimistic bunches.