The Price of Budgetary Uncertainty, Part 2: The Economic Pain
Even while we dodge bullets, we bleed nonetheless.
Congress is currently on Easter recess until next week. Before they left, however, they passed the rest of the FY2024 appropriations bills, narrowly dodging another shutdown. It likely would have lasted at least the rest of that weekend lest they pass another CR. I’m sure they had the text ready, anyway. Technically, since the appropriations were passed after midnight, we had something like 2 hours of lapsed funding. Down to the wire for no good reason, as usual.
This was the second half of the “minibus” appropriations bills, the first of which was passed earlier in March. We now have funding for the government, 5-odd months into the fiscal year. As September ends, we’ll see if there is a repeat for FY2025, or worse. This being an election year makes me lean towards “worse.”
Just after midnight, OMB (Office of Management and Budget) said:
"OMB has ceased shutdown preparations because there is a high degree of confidence that Congress will imminently pass the relevant appropriations and the President will sign the bill on Saturday.”
The implication being that they were watching Congress’ messing around as a way of directing their behavior. The administerial is following the political. Not a great sign, but this has been the case ever since shutdown threats became rather routine. It is emblematic of the US’ overall budgetary uncertainty, and it has costs beyond just those to the public purse in dealing with shutdown procedures, covered in Part I.
Today’s post is Part II in my series on the cost of that chronic uncertainty. Last time, I looked narrowly at the fiscal toll that government shutdowns take. That’s a fraction of the full butcher’s bill. The economic cost of not just shutdowns, but the general chaos that surrounds the entire budget process, is a whole other animal.
I will separate this into sections, as it’s a big topic.
Uncertainty is corrosive to growth.
- Moody’s Analytics, writing on the 2013 shutdown.
We know that government shutdowns are painful. Those that occur may be mild to severe. The longer-term ones (i.e 2018-2019 or 2013) are really painful in relation to their duration. That which we just avoided would have been quite mild, both given that it was a weekend and only half of the appropriation bills.
Direct economic costs are relatively easy to measure. In terms of the basic GDP equation, it means a straightforward reduction in G. Macro 101 also teaches us about the multiplier. In a narrow, short-run sense, when public funds are not being injected into the economy, even via the economic activity foregone by furloughed federal employees, the economy suffers. CBO has estimated that the January 2019 shutdown directly cost us 0.2-0.3% of real GDP.
G simply encompasses government spending, but shutdowns hit the economic water like a cannonball. Delayed investments and projects (recall the train metaphor in Part I), a ton of paperwork and potentially redundant administrative tasks, delayed permitting approval and regulatory hurdles, as well as disrupted contracts are some of the immediate ripples from shutdowns. In the equation, the C, I and X variables are hence also reduced. CBO put it very well:
Underlying those effects on the overall economy are much more significant effects on individual businesses and workers. … Some of those private-sector entities will never recoup that lost income.
You can’t take back wasted time and money. Goldman Sachs estimated at the beginning of FY2024:
A government-wide shutdown would directly reduce growth by around 0.15 percentage point for each week it lasted, or about 0.2 percentage point per week once private sector effects were included.
Not just narrow and short-run. Regarding the 2013 full-shutdown, the Bureau of Economic Analysis estimated that it reduced GDP growth by 0.3% just from lost productivity from furloughed workers, i.e. a fraction of the total lost productivity. In sum, CBO said that real GDP growth during the 2019 partial-shutdown was reduced by 0.4%.
The above also applies to almost shutdowns, more or less synonymous with the practice of continuing resolutions. CRs, unless set for a very long time, are usually reactive, stopgap solutions - a product of incessant political fights and delayed deadlines. Yes, actual shutdowns are worse than near misses, but the ad hoc nature of such habits brings many of the same economic costs. In other words, there is a conceptual distinction between the costs of shutdowns and those of continuing resolutions, although both are tied to the general problem of fiscal uncertainty.
Contractors are an interesting case study, as they straddle the private and public sectors. Many contractors are also active participants in the private market in their respective industries, and thus those activities are impacted by a lack of reliability on the part of their partners in the federal government. When shutdowns do hit, contractors have to keep the business afloat and workers at the ready for whenever funding resumes. From an NBC report:
The result is that companies spend when limited revenue is coming in, causing cash flow problems. "With prior shutdowns, we've had instances where it's particularly hard for small businesses who have difficulty dealing with all this and fronting the money," said Jessica Abrahams, chair of the government contracts practice at law firm Drinker Biddle.
According to a Federal News Network report, there is inconsistency and fog around agencies’ behavior:
“We literally had two companies working side by side in an agency. One got a stop-work order. The other did not. Over time of course some contractors were recalled, and others were not. Often this seemed to be not according to some sort of plan but somewhat random.”
When shutdowns are dodged, and CRs happen instead, problems persist. An excellent study by the IBM Center for the Business of Government, from back during the massive fiscal dysfunction in Obama’s second term, covers a lot of the associated issues. This is a fantastic paper, so forgive me for yet more block quotes:
At one time, many contracts were structured to run concurrent with the fiscal year. As late appropriations became the norm, however, it became standard practice for recurring contracts to be renewed later in the year, under the presumption that by this point it will be clear what level of funding is available.
Sad that this is the case. This is not what they would do if budgeting were predictable and clear.
In addition to delaying operations, agencies were not able to fully compete and award contracts because of the limited time left in the fiscal year at the time the appropriation was enacted. … There are transaction costs associated with each contract, and more contracts means more work. … The chance of an error being made, leading to an audit finding and/or a waste of funds, is greater the more transactions are involved.
The transaction cost point, broadly understood, is key. The lack of stable, long-term contracts between the private sector and government imposes a sort of unforced error, an avoidable deadweight loss. The author puts it mildly:
Inefficiencies that currently exist for federal agencies because of the annual appropriations process could potentially be eliminated in a world of predictable multi-year funding.
Next comes the kicker. There is probably a risk premium that contractors implicitly place on dealing with the federal government. Utterly unnecessary wastes of money through higher prices:
… it seems quite likely that many contractors dealing with the federal government include a “risk premium” in the rate that they charge for contractual services, because they cannot negotiate reliable multi-year commitments without fear of funding interruption. Moreover, this premium likely applies government-wide. … if the eventual level of funding is unknown, or in serious question, it can compromise the grant process by encouraging grant applicants to apply for grants that they are unlikely to receive.
Brien, Letterle and Kantner corroborate this, in their paper examining USMC procurement set against a series of CRs:
This means that years with multiple successive continuing resolutions will require a series of separate contracts to maintain continuity of service.
This may impact the choice to adopt fixed-price vs. cost reimbursement contract designs, and whether the contract is awarded through a competitive vs. sole-source selection procedure [and] … increase the transaction costs for a given purchase and may induce administrators to postpone procurement actions.
… The budgetary politics that lead to continuing resolutions are an example of non-procurement type goals guiding political control over the acquisition system.
Recall that contractors often have business partners and investments in the private market. Most are not simply appendages of the federal government, but actual firms. Sure, they can bid up prices to take advantage of uncertainty, but that’s equally in response to the additional headache that uncertainty creates.
From IBM again:
Companies have a responsibility to shareholders, so they can’t afford to employ people who aren’t working. This means that even the credible threat of a shutdown can create significant problems for contractors. For contractors preparing for a shutdown is not as simple as a division between “essential” and non-essential employees.
Being largely normal private-sector firms, they bear costs not (at least immediately) borne by federal agencies in the case of shutdowns or near-shutdowns. In sum, we have a situation wherein continuing resolutions and threats of shutdowns impose not just higher fiscal costs, but broader economic consequences just from the contracting side of things.
But it’s not just contractors, it’s their business partners, and their business partners, and so on. Economic actors are interconnected. When you have the biggest economic actor of all - the federal government - in frequent limbo, it drags everyone else down.
The federal government’s actual impact on the economy is not merely the G variable in the GDP equation, nor even the fiscal multiplier as normally measured. As mentioned, contractors’ business partners rely on the former’s stability. That’s only part of it, though. The private sector, even outside of the contracting realm, needs the government working reasonably well in order to conduct a large part of its activities. Hence, when things break down, it makes a splash. Licensing, permitting, inspections, leases of government land, grant applications, SBA loans - all get held up. To use a hypothetical, a diner serving Yosemite visitors, for example, may get mauled by a shutdown. Even the threat of one may cause a waste of resources in that diner, not unlike those wasted by federal agencies.
From the WSJ report:
[The owner of ] a retail store and restaurant in Fairfax, Va., said he had to turn to a high-cost, short-term lender because he couldn’t get an SBA line of credit to carry his second location in nearby Leesburg through the slower winter months.
Indeed, in a report on the January 2019 shutdown (remember, tip of the iceberg), CBO said:
CBO’s estimates do not incorporate other, more indirect negative effects of the shutdown, which are more difficult to quantify but were probably becoming more significant as it continued. For example, some businesses could not obtain federal permits and certifications, and others faced reduced access to loans provided by the federal government. Such factors were probably beginning to lead firms to postpone investment and hiring decisions.
Individual Americans’ economic planning is also impacted in analogous ways. Delays or disruptions in cash transfers and employment opportunities are clear examples. Imagine a rancher waiting for the all-clear in their expansion of their property, and the way that this would harm the rancher’s small (potentially very small) business, or an individual worker deciding whether to buy a productive stock or whether to seek a new job.
I’m not actually sure that it’s practically possible to accurately measure this side of the cost to the economy. Again, we have the CBO estimates regarding shutdowns, and there exists research on the costs of CRs to parts of the government, but those are not the same as the subtle losses in efficiency that permeate everywhere when fiscal policy is constantly up in the air.1 Even if you have the temporary predictability of long-term CRs, the amount of funding given to federal agencies does not reflect their actual needs.
Returning to hypotheticals, suppose a year in which there is a surge in applications for permits with the Nuclear Regulatory Commission, but NRC doesn’t have the funding flexibility to process all of these. How far down the line are new projects in nuclear power innovation ultimately pushed? What is wasted or forgone? Repeat this thought experiment among the plethora of federal agencies upon which firms may rely.
I would propose that there is an implicit risk premium on not just contractors, but the bulk of the private sector. “Implicit,” in the sense that many firms without any connection to government, or conscious intent to explicitly respond to fiscal uncertainties, may in reality price this problem in to their business planning. That’s just how markets work at a certain point: vibes. A signal wrapped in an incentive.
It seems that we are stuck with a lack of statistical granularity and completeness, so a lot of this post has been my own theorizing, using the “economic way of thinking.” We ought to remember Bastiat. I could be wrong about the magnitude of the problem, but am quite confident that our wider economy loses out in some significant way from fiscal chaos.
I have only been speaking of the short-medium term, in the context of shutdowns and CRs - regular hurdles the economy faces from the yearly budget process. That’s bad enough for now. Markets usually seem to price in the risk that comes with quarter-to-quarter fiscal disorder, and we plod along. That said, another demon looms over the horizon.
Every year that we do this song and dance, it adds a toll to our long-run economic and geopolitical prospects. Things snowball. I haven’t even touched on the risk that the ballooning sovereign debt poses for the US, nor the political message that Congressional chaos sends to other countries, nor the creeping effects of rising interest rates and the emergent behavior in international financial markets. If some new crisis occurs in the near future, will we be able to emerge unscathed? If not, how scathed?
I will cover all of that in Part III.
I will be doing an extra post on this paper from Moody’s Analytics, which is by far the best attempt I have seen at rigorously studying the problem I have been addressing. This is basically required reading.