Tyler Cowen links to this Matt Yglesias post from yesterday. The “money” shot is this one:

There are two obvious points the author is making here. First is that most “public investment” happens at the state and local level. While I do respect that this is an underappreciated and misunderstood fact, it is our own fault. The federal government would have you believe it IS doing everything, and nationalizing health insurance and educational standards, for example, doesn’t inspire much thought that anything is beyond the reach of the federal government. Furthermore, this tells us nothing about federal influence on investment expenditures. How much of this is dictated by federal tax policy, block grants, mandates and the like? So while the distance between the red and blue lines is large, and should be more widely understood, that in no way tells us a complete picture of what is going on.
The second point of course is that since the end of the Great Recession, state and local public “investment” has “collapsed.” And this seems to be an issue on both the moderate “right” and “left” with Cowen seemingly in agreement with whatever unspoken conclusion Yglesias intends to have us draw from that picture.
Of course, I think I disagree with that notion. First, it would be nice if the two authors who have so many readers would define exactly how FRED is measuring “investment” and to demonstrate what sort of catastrophic public withdrawal we are seeing. Second, why is “consumption” by states and local governments in there and what does this mean? If you search the FRED site for the series Yglesias presents, you find a link to the National Income and Product Accounts, from where this data come, and there is a short data description that comes with it. In there I found:
Government consumption expenditures and gross investment (1–29), the measure of government sector final demand, consists of two major components: Current consumption expenditures by general government and gross investment by both general government and government enterprises. Consumption 19. Own-account production refers to an asset produced by a business or government for its own use. A Guide to the NIPAs 9
expenditures consists of the goods and services that are produced by general government, less sales to other sectors and own-account investment. As producers of nonmarket services, governments generally provide
services to the general public without charge, for example, law enforcement services, national defense services, and elementary and secondary education. The value of government production, that is, government’s gross output, is measured by the cost of inputs: Compensation of employees, CFC (a partial measure of the services of government capital), and in termediate goods and services purchased.20 Therefore, government consumption expenditures is measured as the sum of these costs of production less sales by government of goods and services to other sectors (which are classified as PCE, if purchased by individuals, or as intermediate inputs, if purchased by businesses) and the value of software and construction that are produced by government for its own use (that is, own-account investment, which is classified as part of gross government investment). Gross investment consists of purchases of new structures and of equipment and software by both general government and government enterprises, net purchases of used structures and equipment, and own-account production of structures and of software. Government consumption expenditures and gross investment does not include current transactions of government enterprises, current transfer payments, interest payments, subsidies, or transactions in financial assets and in nonproduced assets such as land.
So the chart above isn’t a chart of “investment” unless we are now entering the post-modern definition of investment as, “anything the government does that I like.” This measure includes the things government delivers to us in the daily course of activities such as the imputed value of police protection, the court system, government schooling and so on. It ALSO includes a measure of what we think traditionally of as investment in infrastructure, buildings and the like. This chart says nothing about the breakout of these things, nor does it, of course, say anything about where the “cliff” is coming from and whether we ought to be worried. What if, for example, the entire decline in “investment” happened to be because we ended the stupid war on drugs and our prison and law enforcement staffing needs fell by billions? Would this be a crisis? Now, we KNOW that did not happen. But it is worth a mention, isn’t it? In a future post, we are going to show you some data from the National Association of State Budget Officers that tries to uncover what exactly has been going on.
As for collapse and a lost decade? Well, public “investment” peaked in the third quarter of 2009 at $1.574 trillion. It is now at a (local) trough of $1.453 trillion. This is an aggregate decline of 7.7 percent over 3.5 years. Large by historical standards? Yes indeed, and it is back to levels last seen in the recession of 2001. But remember we had a government “investment” blowout during the 90s, which appeared to be way above trend, and if we plot a fitted line through a long term series (longer than FRED reports for now, I can dig one up with harder work at a later time), public “investment” is 24% higher in real terms than it was in 1995 when the series from which Yglesias draws began. That’s a real increase of over 1% per year. Remember what we mean by “real.” Is it clear that in real terms public investment (in presumably “public” goods) should be increasing at all?
A final point today that I think is worth making is whether “investment” has fallen off? Certainly during the recession we know it must have, it is really the major driver of business cycles. But if you look at measures of Real Gross Private investment, it seems that private investment since the Great Recession ended has more than compensated for the “loss” of public “investment:”

Private investment is up $600 billion since the trough it hit during the recession, over 4 times larger than the decline in public “investment” over this same time. Again, this says nothing about how well this has been invested, or whether it has been politically driven, or anything like that. And indeed, depending on which tribe you are in, you can cite this as evidence FOR or AGAINST the multiplier. But I digress.
There is a lot more to uncover here and we shall endeavor to do so by examining what, exactly, state budgets have looked like over time, and how their spending has changed,