First, opportunity cost. Resources are finite, governments can not increase the amount of resources it can only redirect them. Before the government can direct resources towards research it must first take resources away from other endeavor.
In war time we do not see the “greatest growth in technology” we see technology advanced are confined to a limited field: weapons. Since weapons see little development outside of war times, major advances occur once war breaks out. Of course, the opportunity cost here is that development of new weapon technologies means that fewer productive technologies are created.
Now we get to the calculation problem. How does the government know how much research is enough and how much is too much. If every person were employed as a scientist, technology could advance at the fastest pace possible, but the trade off would be that no one is working to grow food, build homes, or in factories.
Clearly its possible for there to be too much research so we must find the proper balance between production of technologies and production of consumer goods. That proper balance is determined by the preferences of consumers. And the only way individuals can demonstrate their preferences is through the market.
Any amount of research other than the amount consumers are willing to bear is the wrong amount and forces people to be worse off then they would be if they were left alone.
JonBostwick seems to approach this the “macro” way. Often I do the same.
When discussing R&D qua R&D (as opposed to R&D qua a specific technology or qua libetarianism) I find Fritz Machlup’s “Can There Be Too Much Research?” (Science, 1958) to be a good starting point. The added bonus is that few will question an article in Science out of hand.
Here’s some material on Japan and Korea. It’s mostly focused on protectionism, but the type of protectionism they employed centered around high tech firms.
Jeffrey Tucker uploaded The Concise Guide to Economics to Scribd the other day. I seem to recall a short chapter on the calculation debate (chapter thirty-something) in it with references for further reading, FWIW.