What's the optimum net worth of a charitable organisation/university?

Last night I was chatting with someone who works for a think-tank. I work for Carleton University. We both work for charities. Like households, firms, and governments, charities have income and expenses, and a net worth. Consumer choice theory tells us, at least in principle, how much people should save, and what the optimum net worth of a household should be. Finance theory tells us, at least in principle, what the optimum net worth of a firm should be. Public finance theory tells us, at least in principle, what the optimum net worth of a government should be. (Well, that last statement is actually debatable, but let it pass). I don't know of any theory that tells us, even in principle, what the optimum net worth of a charity should be. What is the net worth, below which it should be saving, and above which it should be dissaving?

The person who works for a think-tank gave an inventory-theoretic answer for his organisation. He said it likes to aim for a net worth equal to x months expenditures, because its income is lumpy and variable, while its expenditures are hard to change quickly. A reserve of x months seemed to be just enough to smooth out its expenditure. That seemed like a fairly good answer to me. But it didn't seem as applicable to a university.

Carleton University has a small "accumulated deficit", a negative net worth as measured by the accountants, though it's only a couple of percent of our annual income and expenses. (Many Canadian universities are in much worse financial shape.) But the accountants have a very strange way of measuring it, because they value our liabilities at roughly market value, and our financial assets at roughly market value, but our real estate assets (land and buildings) are valued at historic cost.

Since we bought the land in the 1940's, and the buildings have been going up ever since, and there's been a lot of inflation since then, the accountants (following proper accounting conventions) understate our net worth. If we sold the whole campus for a market price, then bought it back again, or rented it back, or re-built the whole thing elsewhere, the accountants would presumably record a large accumulated surplus instead of a small accumulated deficit.

It's not easy talking to accountants about these matters. Their lips move, but I can't hear what they're saying. They don't seem to have a theory. Any measure of something (like the CPI) is only valid or invalid with reference to a theory of how you plan to use that measure to make decisions. And since I can't figure out the accountants' underlying decision-theoretic framework, their measures always seem to be arbitrary, or mere conventions.

But then I don't have a theory either, in this case. If I had some sort of theory of what the optimum net worth of a university should be, I would then know, at least in principle, how I should most usefully measure that net worth. But I don't, so I can't really put forward an opposing view of how we ought to measure the net worth of Carleton University.

Ought a university attempt eventually to pay off any loans it used to acquire land and buildings? If so, that means it should aim to have a positive net worth. But if it merely rented the land and buildings, would that mean it should aim for a zero net worth?

I don't have a theory, and that bothers me.

Comments

[JKH]

"The person who works for a think-tank gave an inventory-theoretic answer for his organisation. He said it likes to aim for a net worth equal to x months expenditures, because its income is lumpy and variable, while its expenditures are smooth but hard to change quickly. A reserve of x months seemed to be just enough to smooth out its expenditure. That seemed like a fairly good answer to me."

That's an answer about liquidity, not net worth. You can have negative net worth but be liquid if duration of liabilities > duration of assets. Your think tanker is confusing liquidity reserve with capital.

Just like neoclassicals confuse the central bank account with capital. :)

Would probably help discussion if you presented a model (not necessarily actual) balance sheet showing asset liability structure.

[Nick Rowe]

JKH: That was quick!

I see your point about liquidity. I think the unstated assumption is that the think-tank cannot borrow, and has no assets like real estate, so that liquidity and net worth come to the same thing. But I think you've put your finger on one of the reasons his answer didn't seem applicable to a university.

Here's a massively simplified balance sheet:

Land: historic value $1; current market value $101
New residence: current market value $100; mortgage $100
1 professor who is owed 1 year's sabbatical at $100 salary (he is postponing his sabbatical till next year): liability $100.

Net worth as accountants measure it = ($1 + $100) - ($100 + $100) = -$99

The postponed sabbatical is an interesting case. If the salary were constant, it's like an interest free loan to the university. If the salary were rising at (say) 3%, it's like a loan at 3% interest. But if you argue that the professor's salary is rising at the same rate as his productivity (hmmmm, but never mind), it becomes an interest-free loan again.

[Nick Rowe]

Actually, the sabbatical case is even more interesting from an accounting perspective.

Suppose a prof is due a sabbatical at 100% of salary (it's less, but never mind) in 2007 and 2014 etc. And salary stays constant at $100. And a sabbaticant must be replaced by a term instructor at the same salary (hmm, but never mind).

If the prof asks to postpone, he takes sabbaticals in 2008, 2015, etc.

If the university asks/tells the prof to postpone, he takes sabbaticals in 2998, 2014, etc.

The first case is equivalent to giving the university a permanent interest free loan (assuming he lives forever, or retires at random and is replaced by another prof who does the same thing).

The second case gives the university a 1 year interest free loan of $100.

[Nick Rowe]

Typo: for "2998" read "2008".

[JKH]

I spoke/wrote too soon.

Really, to do a proper financial analysis in most cases, you need a balance sheet, an income statement, and a flow of funds statement. Then you do some projections, etc.

Your model balance sheet is a bit odd, but a couple of observations.

The accounting for the value of the land doesn't matter. If you "write up" the value to $ 100, you write up equity as well. Accounting net worth has improved, but there's no impact on cash position or cash flow absent other information.

Any proposal to "monetize" the land value by sale and lease back would probably have a favorable cash/liquidity position effect, but wouldn't necessarily affect net worth much.

The sabbatical (hidden agenda in the question?) aspect seems like an odd item from an accounting perspective. Seems like that sort of thing could be an income statement item; i.e. pay as you go without recording it prospectively as a balance sheet item. I don't know. Maybe its comparable to an unfunded pension liability, which is a balance sheet item for a company and/or its pension fund.

You probably need to work this problem with more info about income statement, cash flow, etc.

[Frances Woolley]

What makes a charity different from a firm, other than having at some point filled in a bunch of forms convincing Canada Revenue Agency to give it special status? Basically, once you have charitable status, CRA almost never takes it away, because the penalties associated with loss of charitable status are so large that CRA is reluctant to impose them. So there's a big charity business out there.

The optimal strategy for anyone, firm government household or charity, depends upon the objective function.

[Frances Woolley]

For an argument that some universities (at least) have too large a surplus see http://ethicist.blogs.nytimes.com/2009/09/28/should-you-give-to-harvard/

I seem to remember that the
Tsunami relief fund also ended up with a ridiculously large surplus - the victims, unfortunately, all died, so there wasn't much that could be done in the relief department.

But when Carleton students did what I figure was a good thing - proposed that they would stop raising funds for the hugely endowed cystic fibrosis foundation - there was an international outcry. Admittedly they gave a really offensive reason - cystic fibrosis affects white men - when they should have said 'white men need help, and let's give where our dollar will make a difference, e.g. the Ottawa Mission or the Max Keeping foundation that helps at-risk kids participate in sports'.

It's as if we're afraid to question the worth or value of what charities, especially medical charities, do.

[J. Powers]

Nick,

This is a fabulous question. Since charities (or non-profits or NGOs) increasingly have been taking up the slack to fill in doing morally charged work where gov't can't or won't, it seems clear that we need better tools to evaluate their performance. (I'm not saying that gov't _ought_ to do more; I'm merely noting that it seems like there are some important instances where gov't's mandate has been shrinking.)

Your use of the term "theory" seems a little tendentious to me. You seem to mean a set of governing assumptions from which we can make inferences, as well as a set of rules about how to make the inferences. (But at times you simply seem to mean "a model" or "an explanation.") Accountants certainly work within a theory--you said yourself that it has to do with how things fit within a system of plans and decisions--but it's not generally their job to mess with the scaffolding of their intellectual framework. That means all their rules seem arbitrary--from the _accountant's_ point of view. But they're rarely arbitrary overall.

As a strategy for answering your question, I think it might be interesting to look at universities' charters. Their stated missions, which are putatively their raisons d'etre, have lots of numbers attached. How many people to educate? How many profs. will be needed to teach? What other ancillary services can/should be provided? Etc.

Corporations are also--at least theoretically--circumscribed by their charters. They exist at the pleasure of the public, which presumably grants them the right to exist on certain conditions. (I find it hard to believe that "to make me rich" is a compelling reason for the public to grant a charter; some one or several other things must be at stake.) So I suppose the slightly more general question that might be asked is: how is optimum net worth governed by mission?

Does that suggest some ways forward?

[Frances Woolley]

Thinking about this some more... charities are typically funded, at least in part, through donations - the first definition of charity in my dictionary is giving voluntarily to those in need.

So think about a donor's options:
(a) give money to a charity now for that charity to save and invest to do work in the future
(b) keep the money, save and invest it yourself, and then give it to the charity in the future.
The advantage of option (a) is that a charity may have more advantageneous tax treatment on its investment income. The advantage of option (b) is that you don't have to risk the charity wasting your money on, for example, sending you letters asking for more money. And also one can't keep on doing option (b) forever (you know you're middle aged when your alma mater starts writing and suggesting you remember them in your will, as SFU did the other day...)

In North America, especially out west, a lot of universities were given land grants. So rather than give universities tax revenue (which was hard to raise), the universities were given crown land (which was in ample supply) and then given the task of raising funds from their land - which lots of universities (e.g. UBC) have been doing very effectively in recent years! If the government wants to fund universities, but has lots of illiquid assets (e.g. land), then providing universities with an endowment of land rather than a stream of income might be a good strategy.

[Patrick]

Maybe one has to consider the economic value of what the university/non-profit produces. A university doesn't produce a cash flow, but it does produce flow of educated citizens. Those citizens are more productive than they would otherwise have been absent a university education. The benefits of producing educated citizens don't accrue to the university the way profits would accrue to a firm, but it seems to me it should figure into the calculation when trying to figure out what the schools 'net worth' really is.

[Nick Rowe]

Just realised: postponed sabbaticals are a liability to the university in the same sense that currency is a liability to an inflation-targeting central bank ;)

Back on topic.

No hidden agenda, JKH, I'm just back from sabbatical ;).

The simplest model of a university's income and expenditure is this: the revenue per student is fixed exogenously by Queens Park (very close to true). The expenditure per faculty is fixed exogenously by competition from other universities (not so good an assumption). So if income = expenditure, the average class size is determined by the ratio of S per faculty to $ per student. By saving now, increasing net worth, and spending more later, the university faces a trade-off between class sizes today and class sizes in the future.

"The accounting for the value of the land doesn't matter. If you "write up" the value to $ 100, you write up equity as well. Accounting net worth has improved, but there's no impact on cash position or cash flow absent other information."

That's my sense as well. But the Board of Governors (I'm on it) has decided to reduce the accumulated deficit down to zero over the next few years. That may or may not be a good policy, but the way that policy is "framed" (sorry), the real decisions of the university depend on the way the accounting defines the accumulated deficit.

J. Powers: OK, that makes more sense of accounting. Accountants *want* to be atheoretical (from my perspective on theory), because they want a set of numbers that can be useful to many different theories. A beautifully Quixotic quest, I think, but kudos for trying.

Frances: "So think about a donor's options:
(a) give money to a charity now for that charity to save and invest to do work in the future
(b) keep the money, save and invest it yourself, and then give it to the charity in the future."

I like that way of thinking about the problem. But if the charity has the same rate of return on invested funds as the donor, and the charity acts as the donor's agent, I think that means the charity should always aim for a zero net worth, and let the donors decide the timing of spending? Or is there a Modigliani-Miller theorem lying in wait, where donors just undo any saving the charity does, so it's irrelevant?

I like Patrick's approach too, where you just think of the charity as maximising some cost-benefit NPV problem, where the optimal timing of spending should just drop out of the solution as a by-product.

[dlr.myopenid.com]

Hi Nick,

There is a lot here I don't agree with, but I think hopping to the theory might miss a potentially important point about this specific case. I have a lot of trouble believing you that Carleton has a negative net worth, even as mismeasured by historical cost accounting. This is a strange thing to say given that I know nothing about Carleton and you are on the Board. But it seems very unlikely, even if the accounting is fully accruing future benefits liabilities and includes a big fake liability for deferred capital contributions. Remember that an accumulated deficit is not the same as negative net worth. This is especially true for a university where its net worth is not expected to be built up by generating operating surpluses. A small accumulated deficit only means that over time expenses have exceeded revenue; capital of course can come from other sources besides (and before) flowing through revenue. My guess is the stated goal you mention doesn't mean positive net worth, but instead a positive (or zero) accumulated deficit to make up for past "losses." That, of course, can't possibly be a theoretical economic goal so it must be a political goal (or maybe an arbitrary goal). In any case, targeting an accounting construct of course is just a measurement problem and is separate from the theoretical capital structure question you are asking. Apologies if I'm wrong about all that; I admit to pure speculation.

As to the theoretical goal, I agree it is a hard question. But I don't agree that taking a donor-agent approach implies holding zero net worth. Firms take a shareholder-agent approach but don't conclude that zero net worth is optimal. They try to minimize the cost of capital. Clearly, cost of capital is dark matter when it comes to donations. But universities do have real frictions associated with insolvency or near-insolvency emergencies, and presumably donors wouldn't want them to take those risks unnecessarily, whatever their "required return." Financing 100% of assets with liabilities seems certain to create a cost of capital that is higher than necessary. And of course liabilities aren't even income-tax deductible for non-profits. So a pure liability theory of non-profit cap structure seems like a non-starter to me, including under a pure donor-agent approach.