Originally published October 23, 2011 on the New Economic Perspectives blog.
In the previous two weeks we have shown that government budget deficits take the form of net credits to bank reserves at the central bank and as well to the deposit accounts of those who receive net government spending. Normally, this leads to excess reserves that are drained through the offer of government bonds, sold either by the central bank or by the Treasury. Hence, budget deficits normally result in net positive acquisition of Treasuries. But even if they do not, then on government sector ends up with net saving in the form of claims on government.
To put it as simply as possible: government deficit spending creates nongovernment sector saving in the form of domestic currency (cash, reserves, Treasuries). This is because government deficits necessarily mean the government has credited more accounts through its spending than it debited through its taxes.
Remembering the comments on Blog 20 we need to make clear that we are talking about net saving in the domestic currency. The domestic nongovernment sector can also net save in foreign currency assets. And some members of the nongovernment sector can save in the form of claims on other members of the domestic nongovernment sector—but that all nets to zero.
It is now obvious from the previous discussion that the nongovernment savings in the domestic currency cannot pre-exist the budget deficit, so we should not imagine that a government that deficit spends must first approach the nongovernment sector to borrow its savings. Rather, we should recognize that the government spending conceptually comes first–it is accomplished by credits to bank accounts. And finally we recognize that both the resulting budget deficit as well as the nongovernment’s savings of net financial assets (budget surplus)are in this sense residuals and are equal.
As a sidenote (for now) those who claim that the US government must borrow Dollars from thrifty Chinese don’t understand the most basic accounting. The Chinese do not issue Dollars—the US does. Every Dollar the Chinese “lend” to the US came from the US. In reality, the Chinese receive Dollars (reserve credits at the Fed) from their export sales to the US (mostly), then they adjust their portfolios as they buy higher earning Dollar assets (mostly, Treasuries). The US government never borrows from Chinese to “finance” its budget deficit. Actually, the US current account deficit provides Dollar claims to the Chinese, and the US budget deficit ensures these are in the form of “currency” (broadly defined to include cash,reserves, and Treasuries).
More generally, as J.M. Keynes argued, saving is actually a two-step process: given income, how much will be saved; and then given saving, in what form will it beheld. Thus, many who proffer the second objection—that nongovernment portfolio preferences can deviate from government spending plans–have in mind the portfolio preferences (that is, the second step) of the nongovernment sector.
How can we be sure that the budget deficit that generates accumulation of claims on government will be consistent with portfolio preferences, even if the final saving position of the nongovernment sector is consistent with saving desires? The answer is that interest rates (and thus asset prices) adjust to ensure that the nongovernment sector is happy to hold its saving in the existing set of assets. Here we must turn to the role played by government interest-earning debt (“treasuries”, or bills and bonds) to gain an understanding.
For the purposes of this discussion, we can assume that anyone who sold goods and services to government did so voluntarily; we can also assume that any recipient of a government “transfer” payment was happy to receive the deposit. Recipients of government spending then can hold receipts in the form of a bank deposit, can withdraw cash, or can use the deposit to spend on goods, services, or assets.
In the first case, no further portfolio effects occur. In the second case, bank reserves and deposit liabilities are reduced by the same amount (this can generate further actions if it reduces aggregate banking system reserves below desired or required levels—but those are always accommodated by the central bank to the extent that attempts by banks to adjust reserve holdings cause the targeted interest rate to move away from target). In the third case, the deposits shift to the sellers (of goods,services or assets). Only cash withdrawals or repayment of loans can reduce the quantity of bank deposits—otherwise only the names of the account holders change.
Still, these processes can affect prices—of goods, services, and most importantly of assets. If deposits and reserves created by government deficit spending are greater than desired at the aggregate level, then the “shifting of pockets” bids up prices of goods and services and asset prices, lowering interest rates. Modern central banks operate with an overnight interest rate target.
When excess reserves cause banks to bid the actual overnight rate below the target, this triggers an open market sale of government bonds that drains excess reserves. (As discussed in the response to comments last week, we modify this if the target interest rate is zero; or if the central bank pays a support rate below which excess reserves cannot push market rates.)
So the answer to the second objection about inconsistency of portfolio preferences is really quite simple: asset prices/interest rates adjust to ensure that the nongovernment’s portfolio preferences are aligned with the quantity of reserves and deposits that result from government spending—and if the central bank does not want short-term interest rates to move away from its target, it intervenes in the open market.
It is best to think of the net saving of the nongovernment sector as a consequence of the government’s deficit spending—which creates income and savings. These savings cannot pre-exist the deficits, since the net credits by government create the savings. Hence, the savings do not really “finance” the deficits, but rather the deficits create an equal amount of savings.
Finally, the fear that government might “print money” if the supply of finance proves insufficient is exposed as unwarranted. All government spending generates credits to private bank accounts—which could be counted as an increase of the money supply (initially, deposits and reserves go up by an amount equal to the government’s spending).
However, the portfolio preferences of the nongovernment sector will determine how many of the created reserves will be transformed into bonds, and incremental taxes paid will determine how many of the created reserves and deposits will be destroyed.
Next week: we’ll get more deeply into bond sales by government and impacts of budget deficits on interest rates.
