When the government creates the same amount of money and directly spends it on infrastructure as opposed to loaning it to infrastructure providers... what changes other than the fact that the infrastructure becomes cheaper to the public since the provider does not need to pay back the loan?
It's massively less efficient. It's important to note that this is true by definition, not by assumption.
For starters, the provider absolutely does still "pay back" the loan either way. Infrastructure investment isn't free, and if the aggregate value produced by the investment adds up to less than what it cost in the first place, then we would have been better off not doing anything at all.
Secondly, the government isn't able to spend on all types of infrastructure, and when it does, it's almost always less efficient. Public works are only one type of infrastructure spending, and they're pretty much the only one that remains exclusively in the domain of government projects these days.
To give you an example of what "infrastructure" can look like, look at the website we're on. Silicon Valley has a robust network of companies and organizations that both provide capital as intermediaries[0] and increase the ROI of that capital[1]. This wouldn't be possible if there weren't sufficient liquidity
When money is created by the banks, it's created in response to market forces. Banks have to turn an economic profit of zero on the aggregate of all loans they extend, which means that they price them accordingly, and therefore the price of capital converges (in the long run) to the value added by that investment. This makes it difficult (though not impossible) to secure capital for projects that have an inferior risk-adjusted payoff.
By contrast, the government does not create capital in response to market forces. Keynesians would argue that this is the entire point of government economic policy - to smooth out short-term cyclical trends. The problem with "spending" all newly-created money on smoothing short-term cyclical trends is that it leaves zero liquidity allocated to long-term economic growth.
[0] most startups are still funded by banks, at the end of the day, with VCs serving as intermediaries - venture capitalists get their money from LPs, which tend to be institutional investors like banks (or substitutes for institutional investors).
[1] even aside from the money that YC provides a company, YC makes companies more successful (or at least more likely to succeed) through the other, intangible assets it provides.
George Bailey explained this, in one of the greatest movies ever made.[1]
CHARLIE
I'll take mine now.
GEORGE
No, but you . . . you . . . you're thinking of
this place all wrong. As if I had the money back
in a safe. The money's not here. Your money's in
Joe's house . . .
(to one of the men)
. . . right next to yours. And in the Kennedy
house, and Mrs. Macklin's house, and a hundred
others. Why, you're lending them the money to
build, and then, they're going to pay it back
to you as best they can.
I have absolutely no idea what Switzerland will wind up with if they vote for this. But it certainly won't be banking as we know it.
This is known as a bank run, or (when it happens to lots of banks simultaneously) a bank panic. During the Great Depression, banks were frozen, and people were barred from withdrawing money from the bank for a certain period of time. The motivation for doing this was to halt the bank panics that were occurring.
Raising the required reserve ratio has very far-reaching implications. Broadly speaking, it tightens liquidity, making loans more difficult to come by, which decreases investment in infrastructure. This slows economic growth, because it's harder to find capital with which to start businesses, and it's harder for people to obtain money to purchase a home, further their education, etc.
[0] in the aggregate; not everyone would lose 90%, but 90% of aggregate cash assets would be.