Yes, you have it right, except that you were being too generous with credit for government efficiency. ;)
It's not 90%. I'm not sure what the correct number is, but just for transfer programs, where the government takes taxes in and hands out checks to welfare beneficiaries, the efficiency level is below 50% due to the size of the bureaucracy relative to the flow of money being transferred.
For example, on one gov't project in Seattle, we had to have an affirmative action auditor, who audited the affirmative action officers, just to make sure that they were doing their jobs vis a vis handing out contracts to minority-owned firms. In other words, we have bean-counters watching the bean counters--multiple redundant layers of totally useless bureaucracy, that has nothing to do with getting the job done.
BTW, I was a little glib; MOST government spending is non-productive. But the road system is somewhat of an exception; the economy actually uses it. The problem is, that a bureaucrat decides WHICH roads to repair, which to ignore, and where to build new ones, that are not determined by economic efficiency. So the transportation network is productive but not particularly efficient.
The real keys are
* Are consumption and production in balance?
* Is this the best use of the money?
Nobody can consume until someone produces. A certain amount of reserved purchasing power ("savings") must be allocated to FUTURE production, or the system decays.
Let's say that a municipal agency decides that a certain county needs more jobs and more business. So they decide to create something like "Research Triangle Park" in North Carolina--a network of suburban offices and roads connecting them, to try to lure in businesses.
People see the new roads, the new offices, the new jobs, and they are happy and think that this was a good use for the resources. But was it?
All this money is being spent on things nobody knew that they needed, until a government bureaucrat decided they did. Granny, for instance, is using the new roads to get to Bingo at the new elder community center. She's driving more, and burning up more resources than she did before, because someone has enabled her to do so.
Now then, for any kind of resource, there is usually more than one use to which it could be true. This is true not only of the money, but also of the things money can buy.
For example, a piece of land could be a small farm, or it could be a Wal*Mart parking lot, or it could be a public school, or it could be tract housing, or it could be an apartment building, etc. It can not be all these things at the same time; once it is committed to a use, it is not available for other uses. WHAT IS THE BEST USE?
If we already have too many fast-food restaurants, and the ones we already have don't have enough customers, then a fast-food restaurant would be a poor use of the piece of land, especially if there were a desperate need for more apartments because rent has gotten unaffordable.
What if someone decided by pure fiat to put a fast-food restaurant there anyway? So, profitability of restaurants goes down MORE, and there is still a shortage of apartments.
That is exactly what happens when the allocation of resources happens according to government policies, and not according to the price signals (apartment buildings more profitable than fast-food restaurants). Even if your government bureaucrats MEANT WELL, and wanted to put the resources to their OPTIMAL USE, they could not!
The reason they could not is because there is a "combinatorial explosion"--that is, the numbers get unmanageably big--of possible uses for each and every resource. It is impossible for a bureaucrat to accurately figure out how much office space, or how many apartments, or how many shoes, or how many #10 1.5" bolts we need, and when he guesses wrong, it will cause oversupplies in one area, and shortages in another, possibly dooming all uses!
For example, if we use the land to build a fast food restaurant, that means that we'll need to allocate some other resources to it, such as napkins, happy-meal toys, uniforms, etc, that would NOT be allocated if instead of building a fast-food restaurant we built an apartment building. Now, each of THOSE resources (napkins, happy meal toys, uniforms, etc) represent ONE use of MANY potential uses of the resources that went into them. Instead of making more napkins, we might have made more cardboard boxes or stationary or some other wood pulp good instead of napkins.
What happens when allocation of resources gets too centralized, and furthermore, is motivated by other than pricing levels, is that resources get misallocated. We end up with too many fast food restaurants and not enough hamburgers! The misallocations are not only on the big scale but on the small scale, so that you're missing parts to finish partially-completed components. You have a fast-food restaurant, that goes out of business because it could not get happy meal toys at an affordable price, or the manager couldn't afford a new soda machine after the last one broke down. You end up with a lot of waste. The whole restaurant, which was already marginal due to oversupply thereof, goes belly up for lack of a few parts, or in other words over-investment in one area and underinvestment in another.
99.99% of the population can not understand this. They fixate on the business park, the new jobs and paychecks, granny's bingo night, and fail to imagine ALTERNATIVE USES that the money and resources could have been put to which, due to a better allocation algorithm, would NECESSARILY have been more efficiently allocated. If you can understand this problem, you are ahead of the game. :D
