Government Deficits: A Financial View

The Democratic attempts to invoke Ricardian equivalence to rail against the tax cuts and another brouhaha over MMT have prodded me to finally put down my thoughts on fiscal policy. I am not a card-carrying MMTer. I do not share the view of some MMTers that tax revenues do not matter. Fiscal capacity is a real constraint--just ask any emerging market government--and it is related to the productive capacity of the economy and the ability of the government to tax that productive capacity. To be sure, the fiscal capacity of the United States is probably far higher than most people imagine.

Most of the mainstream confusion about fiscal policy and government debt comes from trying to analyze the modern financial economy as a barter exchange economy and assuming that the ultimate purpose is increasing consumption. The modern economy is intrinsically a financial one and the goal of capitalists is to accumulate financial assets. It so happens that western capitalism has channelized the private desire to accumulate into creating material prosperity for the masses. Viewing fiscal policy through the prism of finance can help clear a lot of cobwebs (talk about mixed metaphors).

The Multiplier

Since the great financial crisis and the Fed hitting the zero lower bound, there is a greater recognition that fiscal policy has an active role to play in macro management. Until then, the dominant mainstream and policy view was that monetary policy could adequately deal with macro stabilization.  (Some unreconstructed monetarists still think this is true.) However, recent research within the mainstream paradigm shows that the multiplier is significant not just at the zero lower bound but whenever there is significant slack in the economy.

More important, the multiplier--which is basically an estimate of the flow effects--in my opinion woefully underestimates the long-term dynamic effects of fiscal stimulus. Although in recent years the safe asset shortage issue has brought some recognition that fiscal policy has balance sheet effects, the flow and balance sheet effects are generally not reconciled in a comprehensive framework in the mainstream.

Why the Multiplier is Inadequate

In most recessions, the private sector is not just trying to cut production and clear inventories but also attempting to restore financial health by improving the liquidity position of its balance sheet. This is especially true in the so-called balance sheet recessions. In this context, fiscal policy, apart from stimulating demand contemporaneously, accommodates the private sector's desire to reduce leverage and increase the safe asset proportion of its portfolio. When the government runs deficits, the private sector by definition is running positive financial balances, which helps reduce leverage and improve liquidity.

In a balance sheet recession, large deficits go hand-in-hand with weak aggregate demand as the private sector "uses" the stimulus to pay down debt and increase safe assets. Under such circumstances, the contemporaneous multiplier will be low. However, note what happens over time. As private sector balance sheets delever and become healthy, it sets the stage for a private sector-led recovery. This may take years in some cases. Take for example WWII. The government ran large deficits and the private sector was "forced" to save by wartime restrictions. However, when the war ended, private sector balance sheets were lean, with little debt and a huge amount of liquid assets. They were ready to take the baton, unleashing the postwar boom. Incidentally, an article by Vanguard finds that the level of government is positively correlated with future stock market returns (hat tip @lhamtil). Even if we use state-dependent estimation--which is common today--unless balance sheets are taken into account, the multiplier will tend to underestimate the effects of fiscal policy. In short, the multiplier sells fiscal stimulus short.

Why Does Fiscal Policy Work

The efficacy of fiscal policy is predicated on the private sector treating government debt as an asset on its balance sheet, which stands in stark contrast to the assumption of Ricardian Equivalence (RE). If people at large act on the belief that the government bonds they hold in their portfolios are matched by deferred liabilities, then fiscal policy will be ineffective. In economists' jargon, is public debt outside money or inside money (RE view)?

Let us review the conditions under which Ricardian Equivalence holds: 1) infinitely-lived households, 2) rational expectations, 3) common knowledge (everyone knows the macro structure of the economy, and knows that everyone else knows and so forth). None of these are remotely true. Even within the mainstream literature, it is widely known that Ricardian Equivalence does not hold in overlapping generations models (which is a relaxation of the first assumption). RE also does not hold in the presence of uninformed consumers. Lastly, even with infinitely lived households and rational expectations, if we drop the assumption of common knowledge, RE does not hold (see https://economics.mit.edu/files/12527).

Consumption is not the be all and end all for capitalists. As I have stated earlier, accumulation is an end in itself. Indeed, even using a DSGE framework, an insatiable demand for wealth can lead to all kinds of non-standard outcomes (see http://www.iser.osaka-u.ac.jp/library/dp/2015/DP0946.pdf and https://hal.archives-ouvertes.fr/hal-01211667/document). The beauty of capitalism is that the insatiable demand for wealth does not mean insatiable demand for real resources. In contrast, to feudal societies of the past, capitalist desire for wealth can be accommodated by paper claims. I would go so far as to say that capitalism works because the wealthy do not insist on consuming all their wealth! Of course, people want some assurance that their paper wealth has real value and when they become concerned financial panics ensue.

Be that as it may, as financial wealth based on risk assets grows, the portfolio demand for uncorrelated safe assets also grows. By accommodating the demand for safe assets, fiscal policy makes private sector balance sheets more resilient and less prone to financial instability. In contrast, monetary policy, which works by encouraging leveraging and increasing asset valuations through lower discount rates, exacerbates financial instability inherent in capitalism.

Limits to Fiscal Policy

Pretty much everyone agrees that, if an economy is operating at full capacity with all resources being more or less fully employed, fiscal stimulus will lead to inflation and crowd-out private spending. The clearest example is once again WWII. The need to commandeer resources for the war effort meant high inflation as well as forced saving by the private sector. Private investment was extraordinarily weak. In the 1960s, in contrast, as the private economy was booming, the government decided to finance the Vietnam Ware through deficits. There were no restrictions unlike in WWII and inflation started to take-off.

Outside of the conditions of full employment, what are the limits on fiscal policy? Does government debt have any relation to future tax revenues? Theoretically, as long as GDP growth is higher than the interest rate on public debt, the government does not even need to run primary surpluses for the Debt/GDP ratio to stabilize. In fact, for the majority of the time in U.S. history the average interest on public debt has been lower than GDP growth (see https://www.cbpp.org/research/federal-budget/difference-between-economic-growth-rates-and-treasury-interest-rates).

So, for debt to become unmanageable, somehow interest rates have to rise, which is essentially a bond vigilante type argument. Long-term yields have to rise and the Fed has to validate that by raising rates in order to assuage the bond market. What if the Fed focuses on inflation and refuses to play ball with the bond market? How can we get inflation from people becoming worried about debt sustainability. One possibility is that the dollar collapses as foreigners attempt to dump their dollar holdings. They may also refuse to sell anymore to the U.S. on dollar payments. (This assumes that their economies have strong enough domestic demand to forego the U.S. market, which is not the case right now.) Then we would experience surging inflation as imported goods become more expensive (or unavailable) and we have to ramp up domestic production, thereby pushing us closer to full employment. However, under such a scenario, it is likely that U.S. Treasury receipts will be surging (bracket creep plus rising nominal GDP does wonders for receipts), the deficit will be shrinking, and the debt/GDP will be shrinking because the denominator will be rising fast. The last time we had surging inflation, the 1970s, the debt/GDP declined even though interest rates rose.

Of course, if the government continued to be reckless in the face of surging inflation, then we would have a problem. Are we on that road? Remember in the 1960s, Lyndon Johnson jawboned the Fed because he wanted to fight the Vietnam War without raising taxes and the economy was at full employment. It led to inflation and considerable turmoil but it did not turn us into Zimbabwe. Unless there is a complete collapse in the U.S. institutions and the productive capacity, it is hard to see how debt by itself becomes a problem. Problems with sovereign debt will tend to be a symptom of deeper malaise than the cause of it. In which case, better to focus on those deeper issues of structural breakdown than worry about the debt.

P.S. A lot of recent mainstream literature has findings that are closer to post-Keynesian type intuitions than the New-Keyenesian ones (RBC is not even a serious player in my opinion). For people so inclined, take a look at George Angeletos of MIT. Working within the mainstream modeling paradigm but dropping a few key assumptions, such as common knowledge, he is able to show breakdown of Ricardian Equivalence, positive government debt in steady state, crowding-in from fiscal stimulus, etc.