At one time I worked at a research institute. It had a huge library that was only partially filled. One of the directors wanted to buy every developer their own Safari subscription. The cost was quoted at around $4K/mo IIRC.

I pointed out that it would be far more cost–effective to simple let us request hard copies of whatever books we wanted, and then they would just stay in the library. No one worked remotely at the time.

We ended up getting Safari subscriptions for everyone.

▲

teddyh 19 hours ago | parent | next [-]

It’s capex vs. opex. A large enough company has a fixed budget for both, and for your situation, I assume that the opex budget had the funds, while the capex did not.

▲

thfuran 18 hours ago | parent | next [-]

This separation and other accounting peculiarities like use it or lose it budgeting cause so much inefficiency.

▲

falcor84 18 hours ago | parent | prev [-]

Maybe it's just some peculiarity I'm missing, but wouldn't a smaller capex sum be a pareto improvement over a larger opex? Is there any way in which denying the suggestion was rational?

▲

thfuran 18 hours ago | parent [-]

Yes, spending less money is better than spending more money (though there can be potential tax treatment differences or other complexities depending on exactly how money is spent), but that doesn't mean the person in charge of approving an employee's expense is authorized to approve expenses however they please. If they're given a budget that says they get X money in one category and Y money in another, they can't spend (X-2Z, Y+Z) and can only choose to spend (X,Y) or deny the request. Big organizations often have a lot of inefficiencies due to process.

▲

falcor84 17 hours ago | parent [-]

To be clear, my question was about whether there is a situation where it is rational from a management accounting perspective to define a unit's budget like this.

▲

michaelt 12 hours ago | parent [-]

Imagine you own a factory that makes plastic widgets. The factory has a widget-making machine.

You'll pay rent on the factory and interest on the loan you took out to buy the widget-making machine whether you make one widget or a million. But the amount you'll spend on plastic and machine operator wages goes up the more widgets you want to make.

So you can plot a line, y=mx+c where y is the amount you spend and x is the number of widgets you make. m is the cost of plastic and worker wages, c is the cost of rent and loan interest. (Obviously this is a simplification)

You would carefully monitor the per-unit costs of plastic and machine operator wages - but if you've got a machine that turns 10 cents of plastic into a $1 widget, the more plastic you put in the more profit you make. Only an idiot would try to save money by refusing to buy plastic.

On the other hand, if you need to take out a loan to buy plastic, that's an ominous sign indeed - because the previous batch of widgets should have sold at a profit, giving you all the cash needed to buy the plastic for the next batch.

Business is going great, in fact you're running your widget-making machine at full capacity, and you decide to buy a second one - which is an almighty expense, but definitely a sensible one. If you accounted for this the same way you account for plastic, your accounts would look crazy - you were making decent profit every year, then you suddenly took a 800% loss one year? - so instead this is shown in a separate place in the accounts.

Obviously, you do want to invest in new machines when it's profitable to do so - but if you have to delay the purchase by a few months, it's not the end of the world. And taking out a loan to buy the machine could be reasonable.

So basically operating expenses (plastic) and capital expenses (new machines) are paid for differently, managed differently, and accounted for differently.

Of course if you don't operate a widget factory, the distinctions and the reasons for them might be a lot less clear.

▲

falcor84 11 hours ago | parent [-]

Did you just write a short lesson to explain the difference between capex and opex? Well... thank you, I guess.

But again, the question is: is it ever rational for a functional unit manager to be given a particular maximum opex budget and not be allowed to capitalize a part of that? What benefit would such a restriction offer to the business?

▲

michaelt 10 hours ago | parent [-]

> What benefit would such a restriction offer to the business?

You approve opex budget for the widget department because you know they can turn 10 cents of plastic into $1 of widgets, and you want them doing that in the forthcoming quarter. You're happy for them to spend a nigh-unlimited amount on inputs like plastic, so long as it's on things with a 90% margin.

But you don't want the department head spending their nigh-unlimited budget on things that don't get sold on with a 90% margin. So you don't let them charge the O'Reilly books to the nigh-unlimited part of their budget.

	▲	falcor84 2 hours ago \| parent [-]
		Thanks, I see how this would make sense in general. But it does go against this particular case, no? L&D should probably be budgeted outside the widget COGS anyway, but assuming that the widget department has a single budget, and they're going to be using it for O'Reilly books anyway, is it still somehow rational to pay more for a subscription than less for owning the books? Is it just that that consider the extra accounting effort to be more expensive than dealing with the relatively small inefficiencies?

▲

crystal_revenge 7 hours ago | parent | prev [-]

I've worked in similar scenarios and advocated for the Safari subscription. The most obvious problem with the physical book solution is that not everyone can read the same book at the same time. In my experience it's very common that, because some topic is particularly relevant for the team, many people will want to read the same book. At the same time, you do not want 30 copies of a book that was read by everyone 3 years ago sitting on the shelf.

And, as far as expenses go for a research institute, $4k/mo is very inexpensive.