Paul Krugman poses the following question:

if you believe that a surge in private spending would raise employment — and even the critics agree on that — it’s very hard to explain why a surge of public spending wouldn’t have the same effect.

Well, actually, all things being equal, a surge in private spending would come at the expense of savings, thus decreasing the supply of loanable funds, hence merely shifting employment more toward the later production cycle as opposed to the early phases of production. At least that's the Austrian take.

However, playing on Krugman's turf of Demand-Side Economics, he attempts to answer his own question here:

But one argument I keep reading bugs me: it’s the claim that spending-based stimulus is bad because economic theory tells us that a marginal dollar of private spending is better than a marginal dollar of government spending.

That’s just wrong; it’s a misreading of basic, Econ 101 level, economics.

Yes, the standard theory of consumer choice says that a consumer gains more utility if he or she gets to freely allocate a dollar of spending than if someone else makes the choices: I’d rather buy myself a $10 meal than have you feed me $10 worth of food that you select.

But that’s not what we’re talking about when we talk about stimulus spending: we’re not talking about the government buying consumption goods for the public at large. Instead, we’re talking about spending more on public goods: goods that the private market won’t supply, or at any rate won’t supply in sufficient quantities. things like roads, communication networks, sewage systems, and so on. And every Econ 101 textbook explains that the provision of public goods is a necessary function of government.

However, Krugman is deliberately conflating "Public Works" with Public Goods. Let us recall from Econ 101 to note Public Goods has a technical meaning, specifically denoting those goods that are non-rivaled and non-excludable. As I indicated previously, the "public goods" argument is a dubious one at best. Even--for the sake argument--you allowed a concession of market failure for such goods, the number of such goods are limited at best, and "government failure" in producing such goods often results in "pareto-inefficient outcomes" as well(perhaps worse than the pareto-inefficiencies from "market failure").

For example, consider the notion that Broadband Internet access should be treated as a "public good" and will serve as a "stimulus" in the process. Clearly Broadband Internet access is neither non-rival nor non-divisible and hence not a public good. If you want it, you can pretty much get it anywhere, even in remote areas if you go the wireless route. However, declaring the "internet as a public good" and subsidizing to the extent it effectively becomes non-rivaled and non-excludable good will inevitably invoke a version of the "public airwaves" argument for content-based regulations, leading exactly to the regulation scenarios I posted below. And it should be noted the new Copyright Czar is in large part a government crackdown on P2P distribution, the one killer app that would actually drive greater demand for household broadband access. So give me a break. Finally, it should be pointed out that Japan's government subsidy of NTT's nationwide build-out of fiber-optic lines to homes hasn't done much for their stagnated economy.

Krugman's deliberate conflation of "public works" with "public goods" in this instance to reassure the believers of the scientific basis of the "Stimulus" is a bit amusing. Otherwise, he has been fairly straight forward in advocating an old school production shift toward centrally-planned government demand under auspices of "stimulating" employment, a Keynesian tenet that both predates "public goods" and Public Choice Theory. He's also been more or less equally straight forward that he desires such a shift to be permanent, not temporary, as this succinct US News & World Report blurb points out. Krugman, in the pages of the NY Times, has not been shy in espousing his hope to use crises as basis for creating organs of central planning.

It shouldn't come as a surprise that at the National Press Club, Krugman admitted he has been wishing his entire life to be able to live through a depression. As summarized here, Krugman is merely repeating his thesis from his book, The Return of Depression Economics, that failures on the private demand side of the global economy are resurrecting the "Keynesian Liquidity Trap" that will serve to kill off the Chicago School Monetarist doctrine a la Stagflation of the 70s served to kill off Keynes. Krugman cynically embraced Naomi Klein's Disaster Capitalism absurd thesis that Milton Friedman is somehow at the root of a doctrine to use war and natural disasters as a means to impose "free markets" on people who otherwise would have rejected such outright. Anything to discredit the great Keynesian nemesis. Of course, there's not one iota of such nonsense ever perpetrated by anyone associated with the Chicago School, the Austrian School, or any legit school of libertarianism. However, there is a long word trail left by the likes of Krugman that embraces "crises" as a means to impose permanent Statism on people who would otherwise reject it. Of course, Krugman does not advocate socialism, or the State ownership of the means of production, rather he embraces Corporatism, the old-style Keynesian Corporatism. And he's counting on disaster and crises to achieve it...