What is the optimal mix of debt and taxes, and does it matter?
I believe that absent of extreme cases, it does not matter.
Good financing does not turn a bad project into a good one. If you burn money, it does not matter where the money comes from, it burns either way.
But _current_government_ is smart and only does good projects. Should it sell bonds or raise taxes?
Ricardian equivalence
_current_government_ wants to do a good project, that it cannot finance with already existing cashflow. Everyone agrees that the project should be done, but they cannot agree on the method of financing it. Should the issue new debt or raise further taxes?
Now Robert Barro comes along and says that it does not matter. Whether you use debt or taxes, the impact on consumption and total private wealth is the same. His argument is the following, households know that current debt increases have to be financed by future taxes. To mitigate against those taxes, they save today by buying these bonds that mature when the taxes come due. If the households were instead to pay these savings directly as taxes, they would have the same consumption, for they would not have to save. In effect the same cash flows from the households to the government, either as tax or bond price, but when the bond comes due, the tax is inevitable.
Assumptions
Intergenerational solidarity
An old man interjects, I have no heir, why should I care? Just sell the bond I am not going to have to repay.
Without an heir, why he should care, social norms, and a sense of legacy may suffice. If current households do not save, because they do not consider the future utility, intervene by raising taxes instead.
Intergenerational solidarity may be weaker today, but interest rates are below their long term average, and to be patient requires to care about the future. The assumption of solidarity holds in my books.
Perfect capital markets
I would like to save, but the minimum investment size for the bond is $100.000, objects a young graduate, it is beyond my means. I cannot borrow that much to lend to the government, let there be taxes instead!
Your friendly neighborhood tokenizer or definitely not black solid mineral matter buys a bond and offers fractions of a bond as a token or ETF for a small fee, so that you may lend.
The imperfectness of capital markets is bound only by its limits to arbitrage —if there is money to be made, they find a way to solve the problem. Well regulated, and with cheap information flow, capital markets are perfect enough for Ricardian equivalence to hold.
Uncertainty about future tax obligations
Predicting your taxes is a hard problem with its own profession. But understanding a new tax may be as complicated as projecting a simpler tax system forward into the future. The more important point is the importance of trust into the government when chooses to borrow, but _current_government_ chooses good projects and is certainly trustworthy!
Government spending does not affect private utility
This assumption appears to be absurd, taken for face value. Its intention is to separate the public choice concerns from the other economic questions. Any form of taxation has political implications of who is going to bear the burden, and the same holds true for the treatment of bonds. Even a project that benefits everyone is going to be politically contentious, for everyone desires justice, but defines it differently from anybody else.
To test the Ricardian equivalence empirically, this assumption is needed, and a complex topic on how to correct for it.
Conclusion
How the government gets its financing does not matter in a first best world. The closer a society is to that world, the better of it is. Often, government project reduce that distance somewhat. In my opinion it is more valuable to ask whether the project is worth doing, regardless of how it is financed, and to ensure a high quality public sector by monitoring it to make it trustworthy.
