I've just started an msc in economics and finance and have to write a short essay. the topic is that the cost of capital is by far the most important thing for a business to consider before it invests. of course, I'd agree with this. but are there any arguments against this? my readings seem to show that cost of capital is the only thing that matters. what about for online businesses? or projects largely financed by a government such as the tesla installation in south Australia? is there a scientific name for investment based on desire to innovate rather than make money?
If, by cost of capital, you mean the marginal cost then there is only one specific case where you can ignore it. That is the case where the decision makers do not feel a need to be profit maximizing. There are a couple of real-world, splendid examples of this. Rupert Murdoch and Donald Trump could be thought of as utility maximizers rather than profit maximizers.
For Murdoch, although he is very aware of things such as opportunity costs, it appears that in the past he has made decisions which maximize his power and influence rather than the shareholders' profits. For Trump, of course, his losses are notorious, but if you view his flashiness as a deep-seated need to be known, loved and respected, then his business decisions make perfect sense.
Marginal utility covers these ideas. You don't really need anything more than that. When discussing governmental decisions, however, you do need to consider the more complex case when voting happens. In that case, you pick up all the issues of Arrow's Impossibility Theorem, Principal-Agent issues and, of course, corruption.
Fisher's separation theorem assumes two or more shareholders. When one shareholder holds the majority of the shares, utility maximization is rational rather than profit maximization.
Now if you stray into WACC, then you are wandering into a minefield. There has never been a successful validation study of either the CAPM or WACC. It also doesn't make economic sense. Why would you worry about your historical costs of capital when these are often the equivalent of sunk costs. If you do want to stray in that direction, read the validation paper by Fama and MacBeth that falsifies the underlying construct, the Lucas critique and Mandelbrot's paper "On the Variation of Certain Speculative Prices. I cannot remember the Fama & MacBeth one but it is very famous and you should be able to find it easily.
I believe the closest "scientific" alternative to cost of capital in investment allocation decisions are a group of factors known as "Environmental, Social, Governance" (ESG). ESG broadly encompasses measures of economic, environmental, social, and political sustainability. Proponents often claim that ESG factors complement cost of capital and expected returns as investment selection criteria.
Research on ESG factors fails to concisely demonstrate a strong correlation to market returns. While the lack of decisive data on EGS factor could be seen as a detriment to their use, their general invariance to returns indicates that it might be possible "to do good and well" at the same time. The prevailing argument for this is that an efficient market should provide sustainable remuneration to firms which value add to society.
Reminiscent of ESG, Spitznagel’s “Tao of Capital” indicates that the greatest financial rewards often come to those who do not seek direct financial gain, but rather by indirect intentions (e.g., invent something, add value, increase sustainability, etc).
So while I agree that cost of capital is the most important consideration in investment decisions, there are compelling arguments and evidence that fiscal criteria are not the only driving force behind both capital budgeting decisions as well as outcome fiscal outcomes. Real world investment decisions and outcomes may also be driven by a desire harmonize notions of economic, social, and environmental utility.
Perhaps it is also notable that most financial models embed notions of normative logarithmic utility (I.e., the rate of utility is constant across time), whereas behavioral financial studies clearly show that this does not actually reflect human preferences for consumption and aversion to risk.
Alternatively, hyperbolic (or harmonic) discounting methods may reflect real preferences and outcomes more accurately since they do not assume that:
a. all returns are reinvested and/or reinvested at an equivalent rate of return
b. utility is a logarithmic decaying function
The net result of hyperbolic discounting is that utility drops off sharply in short term and decrease at a decaying rate henceforth. In practice, this would suggest that far off or highly unlikely events should not be as heavily discounted as they are under normative models, and therefore may be seen as more congruous with sustainable investment criteria.