Dollar Discourse Episode 1: Budget Basics, Part I
So, Jeff. What IS a federal budget, exactly?
The federal budget is the sum total of the results of all the taxing and spending decisions made by Congress and the President. In total, those decisions show what policymakers think the priorities of the nation should be – how much of the economy to take via taxation, and from which portions of the individual and business populations, and how much money to spend on which particular policies and programs.
But the only time that the federal budget ever shows up in one single place is when the President submits his proposed budget to Congress in February of each year. A bundle of documents, collectively called the Budget of the United States Government for a particular fiscal year, predicts what the fiscal results of all of the current taxing and spending laws on the books will be during the upcoming budget year, and lists the President’s proposals for new or changed taxing and spending laws for that year, and the justifications for those proposed changes. But Congress rejects those proposals as often as not.
When thinking about what comprises the federal budget, it’s important to keep in mind that the federal government, essentially, keeps two separate sets of books, one for outgoing money and one for incoming money – a Book of Spending, and a Book of Receipts, if you will (though they don’t actually have names). The sum total new entries into each book are then compared on a daily, monthly, and annual basis by the Treasury Department, and the difference between the cash going out the door and recorded in the Book of Spending, minus the cash coming in the door and recorded in the Book of Receipts, is the federal deficit or surplus.
But not every dollar going in or out of the Treasury is recorded in the two books, and there are a few ground rules about which type of transaction is recorded in which book.
Okay, got it. So what’s the Book of Receipts?
First, let’s look at the Book of Receipts. This book keeps a record of all the money coming into the U.S. Treasury based on the exercise of the sovereign power of the government – the authority to send men with guns to your home or office to collect the money if you choose not to pay. The vast majority of these governmental receipts come from taxes of all kinds – income taxes (individual and corporate), payroll taxes like Social Security and Medicare, excise taxes on things like gasoline, tobacco and liquor, estate and gift taxes, and Customs duties. They are all levied based on the sovereign power of the government to raise revenue, and they are all recorded in the Book of Receipts.
The rest of the Book of the Receipts consists of fines levied by federal courts and agencies, regulatory filing fees, the universal service charge on telecommunications companies, and net transfers from the Federal Reserve.
Is every single payment to the US government included in the Book of Receipts?
Not every check written to the U.S. Treasury gets recorded in the Book of Receipts. When the government borrows money, people buy new securities from the Treasury Department. But this, and other means of financing federal debt, are not part of the budget and are not recorded in the Book of Receipts. Nor is the repayment of federal loan principal and interest (there is separate process for keeping track of loans and loan guarantees). For the fiscal year ending September 30, 2021, the net total amount of the Book of Receipts was $4 trillion.
Well, that makes more sense. So if that’s the Book of Receipts, what’s the Book of Spending?
This one seems more straightforward – a record of all the cash disbursed by the Treasury to fulfill the obligations of the federal government. This includes pretty much any government program you care to name, from Aeronautics to Zoology. In addition, while the repayment of principal on the federal debt is not recorded as part of the budget, interest payments on the national debt are indeed recorded in the Book of Spending. (Checks written by Treasury to make loans and loan guarantees are not recorded as spending, as mentioned above.)
Since the founding of the U.S. government in 1789, the Treasury has kept a constant tally of the cash going in and out of the Treasury and reported it on, at least, an annual basis.
So in essence, we’ve been using the same budget concepts since 1789?
Not exactly. While dollars in and dollars out of the Treasury have always been faithfully recorded, our concepts of these things really mean, and how they should be presented, has evolved over time. For example, the federal income tax was established in 1916. For its first five decades or so, if you got a check in the mail for an income tax refund, that check was recorded in the Book of Spending, the same as if the government was paying you a salary or giving you a Social Security benefit of an equivalent amount.
But a blue-ribbon panel in 1967 caused a fundamental rethinking of budget concepts, and they realized it made more sense to record income tax refunds NOT as positive entries in the Book of Spending, but as negative entries in the Book of Receipts. So, today, any income tax refunds that you or I receive, up to the total amount of federal income tax that we paid, are recorded as negative receipts in the Book of Receipts.
(For lower-income individuals who have very low tax burdens, “refundable” tax credits like the Earned Income Tax Credit and the per-child tax credit mean that they can get a tax refund check for more tax than they actually paid. Refunds up to the zero-tax level are recorded as negative receipts, but refunds in excess of total tax paid are still considered spending and are recorded in the Book of Spending. Since the COVID stimulus checks for individuals were treated this way, there were quite a lot of tax refunds recorded as negative spending in 2020 and 2021.)
Similarly, the 1967 Concepts Commission drew a line between money raised through the sovereign power of the government and money raised through “business-like transactions” with the public. When the federal government sells an asset (like land, or the rights to mine or drill on federal land, or the right to use a certain set of electromagnetic spectrum frequencies), those proceeds are recorded in the Book of Spending as negative spending, also called “offsetting collections.” Similarly for rent charged by federal buildings, or the sale of books, maps and charts published by federal agencies.
Also, real user fees charged to the public in exchange for specific services are also recorded in the Book of Spending as negative spending. Certain Medicare premiums designed to mimic private-sector insurance are negative spending, as are TSA aviation security fees, Agriculture Department food inspection fees, National Park admission fees, et cetera.
Are these two books made public?
The Treasury issues daily and monthly statements of both books, and the federal government now runs on a fiscal year that starts October 1 and ends the following September 30, so there is a final Combined Statement of both books published by November of each year. For the fiscal year ending September 30, 2021, in the Book of Spending, the gross amount of spending was $7.5 trillion, but that was offset by $710 billion in offsetting collections, user charges, et cetera, from the public, leaving a net spending total of $6.8 trillion. As mentioned earlier, this number is subtracted from the $4 trillion in the Book of Receipts for that year, leaving a deficit of $2.8 trillion. (Thanks, COVID.)
Thanks COVID, indeed. Are there any other intersections of the two books except for after-the-fact totals?
There is one exception to the rule that the Book of Spending and the Book of Receipts aren’t really connected. That exception is federal trust fund accounts.
I need to take a minute to distinguish a real-world trust fund from what we call trust funds in the federal budget. In the real world, a trust fund is money or some other kind of wealth being transferred from one party to a second party, but held in trust by a third party until some future point in time. In the real world, that third party almost always has a legal, fiduciary responsibility to manage the money safely and securely and to make decisions in the best interest of the second party to whom the money will eventually go.
The federal budget has things that are called trust fund accounts, but they aren’t really trust funds. There is no third party, nor a real second party – the federal government is holding money for itself, and no one has any fiduciary interest.
What a federal trust fund really is, is a LINK BETWEEN THE BOOK OF SPENDING AND THE BOOK OF RECEIPTS. A federal trust fund is a way to segregate a specific set of government receipts in the Book of Receipts and correlate that with the spending from specific accounts in the Book of Spending, over time. It’s a visibility exercise, not a third-party fiduciary relationship.
These non-fiduciary trust fund accounts first appeared in the Wilson Administration with federal civil service retirement and military life insurance trust funds, where the employee contributions were segregated in a trust fund account so that the multi-year receipts could be correlated with the multi-year payouts. The same idea was then used with Social Security starting in 1935 (then called an “Account” and renamed a “Trust Fund” in 1939).
And knowing the difference is important because of the highway trust fund, right?
Yes. Starting in 1956, much federal transportation spending was converted to the trust fund model. The Highway Revenue Act of 1956 provided funding to build the Interstate Highway System by creating a Highway Trust Fund to receive the proceeds of federal motor fuel and trucking excise taxes and to pay the costs of Interstate construction (and the smaller, pre-existing federal-aid highway program).
It is notable that the Trust Fund was only created in 1956 after an earlier attempt to fund the Interstate, through tax increases deposited in general revenues, was voted down by Congress. Road interests, particularly trucking interests, wanted more certainty that their increased tax payments would not be “diverted” to non-highway purposes, and the creation of a trust fund account seemed to satisfy this political need.
So the trust fund model has expanded beyond highways?
Federal support for the air traffic control system and for airport expansion was converted to the trust fund model in 1970, funded by increased excise taxes on aviation users. Maintenance and operation of the Corps of Engineers’ inland waterways system was taken over by an Inland Waterways Trust Fund in 1978, financed by an excise tax on the diesel fuel used to tow barges. A decade of efforts to create a separate trust fund to support mass transit spending resulted in a separate Mass Transit Account of the Highway Trust Fund being created in 1982, supported by increased motor fuel excise taxes. And the dredging of deepwater ports has been funded since 1986 by a Harbor Maintenance Trust Fund, supported by a tax on the value of cargo imported by ship.
So how are all of these trust funds related?
All of these federal trust fund accounts – Social Security, Medicare, the transportation funds, unemployment, Superfund, what have you – have a few things in common. In times when cumulative receipts exceed cumulative spending, the positive balances are invested in U.S. Treasury securities that pay interest that is credited to the trust fund. (Here is where any analogy to a real-world trust fund breaks down – the federal government (“We, the People”) is holding money on behalf of itself, using that money to cover its own borrowing elsewhere, and paying itself interest for the privilege of doing so. If I move cash from my wallet to my front pants pocket in the morning, and then put it back into my wallet when I go to bed at night, does my pocket owe my wallet interest from having borrowed the money all day?)
Second, since 1969, the cash flow and balances of the trust funds have been part of the unified federal budget. Trust fund surpluses decrease the size of the unified federal deficit, and trust fund deficits increase the size of the unified federal deficit. A previous presentation called the “Administrative budget,” in use from 1932 to 1968, excluded trust funds from the federal spending and revenue totals discussed in Congress and by the public.
The Budget Concepts Commission in 1967 recommended including trust funds once again in the budget totals, for several reasons, but the most irrefutable argument was economic. A dollar spent out of a trust fund account contributes just as much to inflation (a big priority of the time) as does a dollar spent out of the general fund. And a tax increase dedicated to a trust fund account puts exactly the same damper on the economy as does a similarly sized tax increase devoted to general revenues.
Has the trust fund model caused any problems?
The fact that the Highway Trust Fund had been effectively off-budget during the heyday of Interstate construction, and then was put back into the budget framework and forced to be judged alongside all other federal programs each year starting in the 1970s, caused political problems in Congress that lasted until fairly recently.
Finally, the problem with a federal trust fund account where no one has any fiduciary responsibility is that, if spending and taxes ever get imbalanced, Congress has to fix it, and Congress doesn’t always do so in a fiscally responsible way. For example, since 1956, the Highway Trust Fund has taken in $1.3 trillion in excise taxes from highway users, but it has spent almost $1.5 trillion during that same period – about $177 billion more than it has taken in.
$32 billion or so of that difference was covered by the interest paid by the federal government to itself for the privilege of holding its own debt – essentially, just printing money. And the other $145 billion in excess spending was covered by ad hoc bailouts by the federal government – special laws passed that require the Treasury Department to create money to record as deposits in the Trust Fund’s page of the Book of Receipts.
Again, it’s basically just printing money. Since the Highway Trust Fund first ran out of cash in September 2008, Congress has provided over $275 billion in such bailout transfers, without ever increasing the excise tax rates that were supposed to make the Trust Fund self-sufficient.
Anyway, the key takeaways from episode one are:
- The Treasury keeps the spending books and the receipt books separately, and the year-end difference between the cash income in the receipts book and the cash outgo in the spending book is the federal deficit or surplus.
- Sale and redemption of the public debt, and the face value of federal loans and loan guarantees, are not recorded in either book but are handled separately. (Interest paid on the public debt is recorded as spending, as is an estimate of the ultimate cost of a loan or loan guarantee to the federal government.)
- Real user fees, and other types of business-type income, are not listed as positive entries in the book of receipts but are instead treated as negative spending in the Book of Spending and called “offsetting collections.”
- Federal trust fund accounts don’t work like real-world trust funds, but they do serve as a link between a specific set of taxes in the book of receipts and a certain set of spending accounts in the book of spending, over multiple years.
Dollar Discourse, Episode 2: Budget Basics, Part II
Greetings, listeners. I’m Jon Hammond – And I’m Jeff Davis – and welcome to Dollar Discourse: Deciphering the Federal Transportation Budget. I’d like to ask (rhetorically, unless you know how to communicate through time) do you know how a federal dollar gets delivered? If you have questions, you’ve come to the right place. Today, Jeff Davis is going to dive right into how exactly federal spending works.
Jeff, how much money did Congress spend in 2021?
The federal government spent a net total $6.8 trillion in the COVID-inflated fiscal year 2021. But many of those dollars going out the door were the result of fiscal decisions made by Congress and the President one year, two years or more before the spending actually took place.
To understand federal spending, it is extremely important to know that there are three stages of the spending process, and we start with the middle one.
The fundamental concept here is the OBLIGATION. This is when, somehow, the federal government gets put on the hook, and a binding legal obligation is created for Uncle Sam to pay somebody else some money. When a federal contract or grant agreement is signed and executed, or a federal official signs a purchase order to buy something, or when a federal employee puts in a day’s work, a legal obligation is created for the government to pay some money.
Is an obligation the only step of the process?
An obligation is the middle step of the spending process. The third and final step is the cash going out the door of the Treasury and recorded in the Book of Spending, and the legal name for these are OUTLAYS. An outlay is defined as money paid by the Treasury to liquidate, or pay off, an obligation. (This is why federal loans and loan guarantees, and buying back principal on the federal debt, are not recorded in the Book of Spending – because they are not paying off federal obligations.)
If it is time for an obligation to be liquidated, and Treasury or another federal agency refuses to make the outlay, the non-federal party can take the federal government to claims court with a very good chance of winning. Once an outlay is due, it isn’t optional.
This is why the famous Gramm-Rudman-Hollings deficit reduction plan of the mid- 1980s failed – because it was triggered by deficits created by outlays, but by the time the outlays came due, it was too late to do anything to stop them.
So if you want to control the volume or speed of spending outlays, you have to stop or slow federal employees from signing the contracts, grant agreements, purchase orders and hiring papers that cause obligations. (More on that later.)
What other steps exist outside of obligations?
Aside from an obligation, there’s also something else required before an outlay can be made. The Constitution states that “No Money shall be drawn from the Treasury, but in Consequence of Appropriations made by Law.”
Well, that doesn’t seem too controversial.
This was a big problem in the 19th Century, when the nation was expanding, and communication between Washington D.C. and the farther reaches of the land were slow and unreliable. Federal officials in the field would often run up obligations as they saw fit, forcing Congress to enact new appropriations so that the “deficiency” could be paid. Congress got so fed up with being forced to make appropriations to pay off obligations that they didn’t ask for that they passed a series of laws starting in 1873 now known as the AntiDeficiency Act.
The AntiDeficiency Act makes it a felony, punishable by up to 2 years in prison, for a federal employee to create an obligation or make an outlay without getting prior permission in law.
These “permission slips” to enter into obligations are called BUDGET AUTHORITY and are the first step in the three-step process.
Budget Authority, Obligation, Outlay. One, Two, Three, and always in that order. Here’s an example:
One. Congress passes a law creating $5 million in budget authority to install an electric vehicle charging network all around the White House.
Two. A federal employee legally obligates almost all of that money by signing a $4.8 million contract with a private firm to purchase and install the system.
Three. As the key deliverables schedule in the contract are met, Treasury writes a series of checks to the private contractor making outlays to liquidate the obligation.
How does Congress confer “budget authority?”
Budget authority now comes in many forms. The traditional form is an appropriation. These date all the way back to the First Congress – in September 1789, Congress enacted the first appropriations bill, appropriating the tremendous sum of $639 thousand to fund the entire federal government for a year. ($216 thousand for civilian expenses $137 thousand for the War Department, $90 thousand to pay old bills and $96 thousand for veterans benefits, if you must know.)
Congress passes a new round of appropriations bills providing new budget authority every year, and most of this budget authority expires, and is no longer available for obligation, at the end of the fiscal year. But Congress is free to make appropriations last longer than that (except for Army appropriations, which the Constitution says cannot exceed two years). Some annual appropriations, especially for capital programs like infrastructure, are available “until expended” which means forever.
Congress can also write appropriations into permanent law, and increasingly did so in the 20th Century. Federal pension programs, Social Security, Medicare, Unemployment, and most other entitlements have permanent appropriations or other permanent budget authority in statute. Budget authority is automatically created in whatever amount is necessary to fulfill the parameters of the program.
There are other kinds of budget authority that don’t involve appropriations. Congress can create what is called “borrowing authority”, which is budget authority to spend borrowed money. They can create budget authority to spend offsetting collections and fees, either in an appropriations bill or outside it. Recently, Congress even passed a law creating budget authority allowing the TSA to collect and spend the spare change that people leave behind at airport security checkpoints. (In 2021, this was just under $653,687.64 by the way.)
Sometimes it’s enough for Congress just to give an order. In the immediate aftermath of 9/11, and the collapse of air travel, Congress passed a law that simply said, quote, “Notwithstanding any other provision of law, the President shall take the following actions: … Compensate air carriers in an aggregate amount of $5 billion”. Because Congress gave the President a direct order to give somebody a specific amount of money, that law was budget authority.
Does budget authority ever differ for transportation?
There is a specific kind of budget authority used in many transportation programs, and behind that there is a story. After the great expansion of government that accompanied the Civil War, Congress broke up the responsibility for making appropriations and gave it to numerous committees in the House and Senate. Many committees could write their own appropriations for their programs, and this included the committees that created the early federal-aid highway program.
One hundred years ago, in the early 1920s, the House and Senate voted to consolidate all of the authority to make annual appropriations in the House and Senate Appropriations Committees, but the House and Senate did so two years apart, so for a brief time, the House Appropriations Committee was fighting with the Senate Post Roads Committee over how to fund the highway program.
What they came up with was contract authority – a permission slip to sign a contract to build a road for which the federal government promised to pay a fixed share of the cost, which created a legally binding obligation for an appropriation to be made when the bill came due.
This is how most federal highway, mass transit, and airport funding is provided today. The authorizing committee (House Transportation and Infrastructure, and three different Senate committees) pass multi-year laws providing contract authority, which is a form of budget authority. Most of this money goes to state and local government in the form of grants, and then the state or local partners negotiate and sign a project agreement with USDOT, which creates the obligation.
As part of the 1974 budget reforms, Congress restricted contract authority so that the House and Senate are only supposed to consider bills providing new contract authority if they are drawn on a trust fund – not the general fund. So highway and mass transit authority comes from the Highway Trust Fund and airport improvement contract authority comes from the Airport and Airway Trust Fund.
But it’s still 1-2-3: Budget Authority (contract authority in this case), Obligation, Outlay.
Seems pretty straightforward, but this is Congress, after all. How does the 1-2-3 process work in practice?
Sometimes the 1-2-3 process can go very quickly. Social Security has permanent budget authority written in law. Grandma is legally entitled to a check on the first of every month that she stays alive, so on the day of the month corresponding to her birthday, the budget authority springs into existence and an obligation is created on the same day, and then an outlay to write her a check is made to liquidate that obligation on the following Wednesday. Entitlement programs, and programs that are mostly salaries and benefits, have a very quick 1-2-3 cycle. For example, FAA Operations is mostly the salaries and expenses of air traffic controllers and safety inspectors, and 89 percent of its new budget authority, on average, spends out and becomes outlays in the first year.
Capital programs, by contrast, can take a long time to “spend out” through the 1-2-3 cycle. Say that Congress appropriates $2 billion in budget authority in 2022 to build a new Interstate bridge over the Ohio River. It may take a year or more to negotiate and sign the contract and for the state partners to get their own money together, and then it may take several more years for the bridge to be built. The final outlays could come up to a decade after the budget authority was provided (more, if there are lawsuits or environmental permitting delays).
Does the speed of the process ever seriously affect the amount of federal spending every year?
The White House estimated that if all of its budget proposals were agreed to by Congress, in 2023 there would be $5.8 billion in net federal outlays. $1.4 billion of that would be from budget authority enacted in prior years that had not yet been spent.
But no matter the time period involved, it’s still 1-2-3, budget authority, obligation, outlays, always in that order.
And the process is still governed in part by the AntiDeficiency Act, right?
Earlier, we mentioned the AntiDeficiency Act, which makes it a felony to enter the federal government into an obligation without the budget authority to do so. (No one has ever gone to jail for this, but in practice, an ADA violation means that the department head has to write groveling letters to the President and the Speaker explaining why it will never happen again, and the Appropriations Committees find creative ways to punish the budgets of violators.)
The AntiDeficiency Act is also why we have periodic government shutdowns. At the start of a new fiscal year, on October 1, most of the regular budget authority from the previous fiscal year (the annual appropriations) expire. In the absence of new budget authority to pay their salaries, the AntiDeficiency Act makes it a crime for federal employees to be allowed to show up for work, since the act of working creates an obligation to be paid for that work, and since the AntiDeficiency Act also prohibits the acceptance of voluntary services. Hence the shutdowns.
So, ultimately, the reason we have government shutdowns is because of a single law passed in 1873?
But there is another federal law that is relevant to this process.
After World War Two, Presidents grew more eager to stand up to Congress by refusing to obligate, or make available for obligation, budget authority that had been enacted for programs that the President though were unwise. This process was called “impoundment” because the President was putting the money into impound, as it were. This reached its peak under Presidents Johnson and Nixon, when a third of each year’s new highway contract authority was routinely impounded, and Nixon also tried to impound almost all of a new Clean Water Act grant program.
Congress responded in summer 1974, at Nixon’s politically weakest point, by ramming the Impoundment Control Act down his throat by a veto-proof majority. Ever since, it has been illegal for a President to refuse to make budget authority available for obligation under whatever rules Congress set up for that particular program. Even if a President hates a new project or program and thinks it is a terrible idea, he or she must still let the budget authority go through the obligation and outlay steps.
How does this manifest after the Nixon Administration?
This is a big reason why the Highway Trust Fund ran out of money in 2008 and keeps running out of money – because even though the Administration sees it coming, they are powerless to stop it. Federal courts issued rulings in the 1970s under the Impoundment Control Act that prevent the Department of Transportation from holding back highway and transit contract authority from obligation, or slowing down the obligation rate in any way, even though everyone knows there’s not enough money in the Trust Fund to pay those outlays when the bills come due. DOT is just another observer of the crisis and can’t take any direct steps to prevent the next default from coming. Only Congress can do that.
KEY TAKEAWAYS FROM EPISODE 2
3 steps of the spending process: Budget Authority, Obligation, and Outlay. Always in that order, you can never skip a step.
There are many forms of budget authority, mostly appropriations (annual, multi-year, or permanent). Contract authority from transportation trust funds is another form of budget authority.
It is against the law to enter the government into an obligation in the absence of budget authority, and it is against the law to withhold budget authority from obligation.
Dollar Discourse Episode 3: Budget Basics, Part III
Greetings, listeners! I’m Jon Hammond ~And I’m Jeff Davis~ and welcome to Dollar Discourse: Deciphering the Federal Transportation Budget. In our last episode, Jeff Davis tackled the basics of federal budgets. But much like the omnipotent, irate librarian you remember from high school, you might be wondering how exactly Congress keeps track of all of this. Have no fear – Jeff Davis has you covered!
With that in mind, how has Congress recorded budgets, Jeff?
Treasury’s system of record-keeping started back when the only form of budget authority were annual appropriations bills. It is important to understand that today’s appropriations bills are written in the exact same way they were 200 years ago – unnumbered paragraphs of text, each one with a summary title header above it, with each paragraph listing the amount of money being appropriated, to whom, the purpose of the appropriation, and any other restrictions or clarifications.
These paragraphs and their section headers stayed largely the same year after year (only the amounts and conditions would change), so Treasury started creating individual spending accounts for each recurring paragraph of each appropriations bill.
And it’s been pretty straight forward ever since?
The result isn’t always logical – the Highway Trust Fund portion of the entire federal-aid highway program is one appropriations paragraph each year, so it’s one big Treasury account – $58 billion in budget authority this year, one of the biggest in government. On the other end, there are some very tiny budget accounts – at USDOT, there are currently 7 accounts with $10 million or less per year. And sometimes, the Appropriations Committees will make an appropriation of $499 thousand, or a bit less, because they know that those amounts get rounded down to zero in the budget presentation computer systems.
Can only the Appropriations Committees create budget accounts?
Budget accounts can be created in other ways – sometimes Congress will pass a non- appropriations law saying “there is created, on the books of the Treasury, an a separate fund or account entitled blah-blah-blah”.
If an agency or department is big enough, Congress will usually try to make separate accounts for separate kinds of activity. Specifically, the trends are to make separate accounts for an agency’s operating expenses versus its capital programs, in part because of the vastly different spending rates. Research and development activities are also often given their own separate budget account.
That’s the case of pretty much every federal agency, right?
The Homeland Security Department recently switched over to a common account structure for all of its components – each component (including TSA and the Coast Guard) now has an Operations and Support account, a capital account, and an R&D account if they have any R&D activities. The Coast Guard also has a fourth account for retiree pay because it is the only DHS agency whose employees aren’t part of the regular civil service retirement plan.
Federal subsidies for Amtrak used to be structured this way – split between two accounts in the annual appropriations bill, one for Operating Subsidies and the other for Capital and Debt Service. But Congress changed things up in 2015, so that today, the two accounts are Northeast Corridor and National Network. This was done to prevent Amtrak from using its operating surplus on Northeast Corridor trains to mask the operating deficit on long-distance trains and instead apply the profits from the Acela business traveler types towards the tremendous capital needs of the Northeast Corridor.
If usage varies so much, what even use the accounts system?
Accounts are important because, generally speaking, federal agencies can’t transfer money from one account to another. Congress would have to change a law to allow that. Some appropriations bills give some agencies some leeway here to transfer up to, say, 5 percent between accounts, but such permission has to be given specifically by Congress.
Accounts are also important because they are the building blocks of the federal budget. They are atoms, if you will – the smallest unit that can’t be broken up. So adding up accounts is how we answer the question – just how much does the federal government spend on X?
For the first century-plus of the Republic, this question was only answerable by how much money a particular Cabinet Department or independent agency was spending. But the big Cabinet agencies developed, over time, into conglomerates, performing a lot of different missions. For example, the Agriculture Department doesn’t just help farmers – it runs the Forest Service, runs what used to be called the Food Stamp program and other nutrition entitlements, finances electricity, telephone, and broadband access to rural areas, and performs a host of other duties as well.
If you ask the average citizen what the U.S. Energy Department does, most would probably answer something about energy research or availability, and indeed, they do a lot of that. But last year, almost two-thirds of the Energy Department’s total budget went towards manufacturing nuclear bombs for the Air Force and Navy to put on their missiles and in their bombers.
So, the need to have another way to sum up government spending, other than by department jurisdiction, has been apparent for a century now, since the first centralized budget was submitted in December 1921. The next year, in the second-ever budget, the Budget Bureau broke down all federal spending into four types – military, civil, non- functional (debt interest and tax refunds), and the general overhead from Congress, the President, and the Judiciary.
These categories were reached by adding up budget accounts, so as to break up the Department and agency divisions. Each budget account was analyzed and was classified by the fundamental purpose, or function, of government that the account was trying to serve.
What did the classification system look like in 1921?
That first classification 100 years ago broke down civil spending into 20 different functions. In addition to functions like foreign relations, law enforcement, agriculture, et cetera, there were two separate transportation functions: “Promotion, regulation, and operation of marine transportation” and “Promotion and regulation of land transportation.” (At the time, the government was spending almost 20 times as much on marine transportation as land transportation, because of the subsidized shipbuilding boom started during World War I.)
But in that initial classification, there was a separate functional category for public works of all types. The traditional thinking in the U.S. was that government construction of anything permanent, be it a road, canal, hospital, national park visitors center, or courthouse, was all the same kind of thing. Every major city had all of its various construction activities centered in a Department of Public Works, and although the federal government did not, the thinking was that building of public works was a function of government no matter what ultimate purpose a public work served.
So, even though the federal-aid highway program was already underway by this point, its spending was classified as a public work, not transportation. Transportation, then, was an activity carried out by private entities using some kind of infrastructure, whether privately or publicly built.
If it already seemed a little bizarre in 1921, how long did the first iteration last?
This system stayed the same through the 1932 budget, when “promotion and regulation of aerial transportation” was added as a third transportation function. But Franklin Roosevelt decided to dispose of the functional classification system and come up with a new one that separated Depression “Recovery and Relief” progams from the regular government, which was treated as an afterthought.
Under President Truman, the Budget Bureau finally adopted a functional classification system akin to what we have today. Each spending account was analyzed yet again for the fundamental purpose it served, and assigned to a subfunction of government, and those subfunctions were grouped together into functions. Each subfunction had a three- digit number code, ending in a number other than zero, and each function ended in a zero.
For example, in that first classification system, in 1948, there were 14 different functions, starting with function 050, national defense, and then going up to 700, Tax refunds.
At the time, the government was trying to concentrate federal transportation agencies under the Commerce Department. Transportation was part of a larger function called “Transportation and Communication” (at the time, numbered function 450). There were nine subfunctions, six for transportation and three for communication. Starting with 451, we had promotion of merchant marine, then provision of navigation aids and facilities (Coast Guard, lighthouse service), provision of highways, promotion of aviation including provision of airways and airports, regulation of transportation, and other services to transportation.
And there was no section for public works?
Importantly, there was no more public works function. All public works accounts were classified as to the ultimate function that the public work would serve.
As the years went by, new functions of government propped up. Aeronautics research, and its offshoot of space travel, used to be in the aerial transportation subfunction, but after the post-Sputnik space race started, a new budget function for science and space was created. Similarly, after the OPEC oil embargo and the creation of a Department of Energy, a budget function for energy was created in the 1970s. Also, Social Security and Medicare grew so large that they were given their own budget function numbers.
So what do the categories look like today?
Today, there are 19 functional categories of spending into which all federal spending dollars are classified. Each function has a three-digit number ending in zero.
- 050 National defense
- 150 International affairs
- 250 General science, space, and technology
- 270 Energy
- 300 Natural resources and environment
- 350 Agriculture
- 370 Commerce and housing credit
- 400 Transportation
- 450 Community and regional development
- 500 Education, training, employment, and social services
- 550 Health
- 570 Medicare
- 600 Income security
- 650 Social Security
- 700 Veterans benefits and services
- 750 Administration of justice
- 800 General government
- 900 Net interest
- 950 Undistributed offsetting receipts
By nature, using the account as the indivisible unit means there will be some inconsistencies. For example, the Army Corps of Engineers has, for almost 200 years, had a transportation mission (maintaining the navigability of rivers and harbors), and it also has non-transportation functions like flood control and, lately, environmental restoration. Until the mid-1950s, the navigation program had its own appropriations account and was classified as transportation. But then, Congress redid the appropriations structure so that the Corps now had separate construction and operating accounts, no matter what kind of project was being funded. And since, at the time, transportation projects represented less than half of the funding in each account, today, none of the Corps water program is recorded as part of the transportation budget function. It’s all under the “Natural Resources and Environment” function.
Does this problem extend to any other federal agency?
There are similar problems at the Coast Guard. The Guard has always been hard to classify – in wartime, it moves to the Pentagon and is part of national defense. Most of its mission is transportation-related – safety regulation, and keeping shipping lanes open, but it also does a lot of law enforcement (intercepting drug shipments on the high seas, for example), and does defense work even in peacetime.
After 9/11 and especially after the creation of the Department of Homeland Security, there was a lot of discussion about creating a new Homeland Security functional category, but nothing ever came of it, so the Coast Guard and TSA are still classified as all transportation, not law enforcement or national defense. (A few hundred million of Coast Guard operations for Iraq/Afghanistan/GWOT are declared to be defense in the text of the appropriations law each year.)
What has been learned about categorization over the years?
There can be serious debate about how to classify a program, and serious consequences. For example: mass transit. The original federal urban mass transit program started at what is now the Department of Housing and Urban Development in the early 1960s and was assigned to the budget subfunction for “urban renewal and community facilities” since it was believed that mass transit was more closely tied to urban planning and better urban land use than to transportation, the federal focus of which had always been interstate in nature.
But the new Department of Transportation won a tug-of-war with HUD in 1968 and took the transit program from them. As part of that re-evaluation, mass transit spending was reclassified as “ground transportation” starting in the fiscal 1970 budget. As the decades went by, mass transit systems became more integrated into the interstate transportation network, but the federal role in land use planning, and in leveraging transportation facilities to help force better land use, stalled.
Has USDOT ever thought about revisiting that decision?
The Transportation Secretary who successfully took transit away from HUD in 1968, Alan Boyd, told me decades later that he now thought it had been a mistake, and neither transit nor the federal role in urban development had been as successful as they could have been if they had maintained common management and a common mission.
Transportation was given its own budget function all to itself in 1970, numbered 400, and from then until 9/11, we had it pretty cozy. Transportation had its own budget function, its own Cabinet Department, and its own annual appropriations bill. There was no TSA yet, and the Coast Guard was still at DOT. The transportation budget function, the transportation Cabinet department, and the transportation appropriations bill had almost complete overlap. (Today, Transportation shares an appropriations bill with Housing and Urban Development.)
What do the transportation categories look like today?
Today, there are four transportation subfunctions: ground transportation (highways, mass transit and rail), air (aviation plus the NASA aeronautics research account), water (Coast Guard, maritime, and the St Lawrence Seaway), and other transportation (pipelines, multimodal programs, and general department oversight and auditing).
Pre-COVID and pre-IIJA, the transportation budget function got a little under 3 percent of total federal spending each year. 84 percent of that was at USDOT, 15 percent at Homeland Security, and 1 percent elsewhere.
Every year, if you want to see the entire federal budget, in a little detail, in one table, the Analytical Perspectives volume of the budget allows you two to choose from. One is called “Budget Authority and Outlays by Function, Category, and Program, and then the next one (a much longer one) shows the same overall totals but is called “Federal Budget by Agency and Account.” (Both tables add up to $7.1 trillion in budget authority and
$6.8 trillion in outlays in 2021, for example, but the functional table takes 75 pages to get there and the account table takes 475 pages to get there.
So what is spent on transportation as we speak?
In the COVID-inflated year of 2021, the amount of money spent by the federal government on the transportation budget function was a net $154 billion. Of that amount, a little more than two-thirds was spent by the Department of Transportation. This isn’t just because of the normal Coast Guard and TSA spending but because several transportation-focused COVID relief programs were set up at the Treasury Department. Now that those are almost completely spent out, we should soon return to the norm of USDOT spending close to 90 percent of the federal transportation budget.
Dollar Discourse, Episode 4: Budget Basics, Part IV
Greetings, Listeners! I’m Jon Hammond *And I’m Jeff Davis* and welcome to Dollar Discourse: Deciphering the Federal Transportation Budget. If you’ve listened to the last few episodes and wondered how to make *cents* of how the federal budget got to where it is today, then you’ve come to the right place! In this episode, Jeff Davis is going to dive right into how we got to where we are. With that in mind – With that in mind – what else do we need to know?
There is a prayer written by Reinhold Neibuhr that is very popular in 12-step programs. It goes like this: God, grant me the serenity to accept the things I cannot change, courage to change the things I can, and wisdom to know the difference.
I’m a big fan of 20th century theologians myself Jeff, so how does this relate to the federal budget?
In the federal budget, all spending is divided into two categories – spending where the annual level can be changed in the annual appropriations process, and spending where the level cannot be changed in the annual appropriations process.
These two categories have names – the type of spending where Congress has the discretion to change things up in the annual appropriations bills is called DISCRETIONARY spending, and the type of spending where the appropriators’ hands are tied is called MANDATORY spending.
And that’s always been the case?
In the beginning (from the First Congress in 1789), all federal spending was discretionary. But in 1818, Congress enacted a law declaring that all Revolutionary War veterans who needed it were entitled to monthly pensions ($20 per month for officers and $8 per month for enlisted men) for life. The law actually used the word “entitled” to refer to the former soldiers who were to receive these benefits.
This was the first “entitlement” created by Congress, where a legally binding obligation was placed on future Congresses, which had to come along later and make the appropriations to fulfill those promises.
That must mean that entitlements have been pretty common, right?
Other than Revolutionary War (and later Civil War) pensions, there weren’t any other entitlements until the early 20th century and the creation of civil service retirement benefits (about the same time as World War I pensions). And, at around that time, Congress created, in 1922, a new way to fund the highway program. Called contract authority, this authorized states to incur obligations for matching federal highway funds before an appropriation was made, tying the hands of the Appropriations Committees and forcing them to make appropriations fulfilling the contracts.
What happens if the Appropriations Committees don’t fulfill one of those legally binding promises?
Good question. There is, in section 1304 of title 31, United States Code, something called the “permanent judgment appropriation.” Any time a person or company sues the federal government for breach of contract, or because they feel they are entitled to money, and they win and receive a judgment, money is automatically appropriated to pay the judgment (with interest, if necessary).
Is that a fairly large number, then?
As you can imagine, this varies widely by year – in 2021, it paid $8.2 billion in judgments, but in 2022 that dropped to $2.8 billion for some reason.
Anyway, all of these entitlements and other forms of non-appropriated spending were lumped together and called “uncontrollable spending” by the appropriators, who were powerless against it. The only way to cut that kind of spending was to amend underlying law, which the appropriations committees aren’t supposed to do in their annual spending bills.
Then came Franklin Roosevelt, who created Social Security and unemployment insurance. Then came Lyndon Johnson with Medicare, Medicaid, and the War on Poverty programs. Before LBJ, annual outlays from the uncontrollable programs were still less than one-third of total federal spending. By 1974, the uncontrollable share had increased to half, and it has steadily increased ever since, to the point that in the last two COVID-inflated years of 2020 and 2021, the uncontrollable programs represented at least three-fourths of total federal spending.
Since the 1980s, uncontrollable spending has been formally known as MANDATORY spending while the spending that is still up to the Appropriations Committees’ control is called DISCRETIONARY spending. Legally, the latter means that each year, Congress has the discretion to reduce or eliminate funds to any account without exposing the federal government to legal liability.
What else counts as mandatory spending?
The interest paid on the federal debt is part of mandatory spending, technically, but it is often listed as a separate category, because it has an extra degree of uncontrollability built into it. In order to control net interest, first you have to stop running annual deficits, which requires some combination of mandatory spending cuts, discretionary spending cuts, and tax increases. And then, you have to hope that market interest rates go down rather than up when it is time to roll over your existing debt, which is even more out of the hands of Congress.
Remember, the combination of large deficits and high interest rates in the 1980s at one point had net interest running as high as 15 percent of total federal outlays by the early 1990s. The last few years of near-zero interest rates pushed this back down to around 5 percent of total spending, but that is going to increase in the next few years now that interest rates are on the way up again.
The two basic spending categories of MANDATORY and DISCRETIONARY are important to remember, because there are separate ways of measuring changes to each one, and separate ways of controlling each one.
How can Congress measure and control changes to mandatory and discretionary spending?
Let’s start with measuring changes. For the discretionary programs controlled in the annual appropriations bills, it’s pretty simple. You look at the text of last year’s bill and see how much money you spent on each account, and within that account, each program, project or activity delineated by Congress. Then the Administration submits its budget request for each account and tries to demonstrate how much of a change (almost always an increase) is needed, and how much of that increase is for payroll vs program, et cetera.
Congress generally compares each discretionary amount to the amount enacted in the prior year. If you provide more, it’s considered an increase, and if you provide less than last year, it’s a cut. This is the common sense way to do it.
And that’s the same for mandatory programs?
For mandatory programs, it’s different. In order to measure proposed changes, budget scorekeepers have to generate what is called a BASELINE. The amount that Social Security spent in 2021 doesn’t tell you much about what it is going to spend in 2022 because you need to know how many people are going to turn 65 this year, how much money they have paid into the system, how many current beneficiaries are expected to die this year by the actuaries and how much money that will save, and how much the inflation-based cost of living adjustment will be.
A budget baseline models all of those things and tries to estimate, under current law, how much those programs will spend in a given year. These are predicated on economic assumptions – you have to predict an unemployment rate to get a baseline spending number for unemployment insurance, for example, and we already discussed how you have to model interest rates to predict future net interest spending.
It gets even more complicated when you try and model health care costs – not only do you need demographic modeling, but you also have to try and judge how fast hospitals, doctors, and pharma companies are going to raise their rates.
For these mandatory programs, if you propose to make a change that will affect spending in the upcoming year, the dollar amount of the change is not measured against how much money the program spent about last year, it is measured against what the baseline projection for the upcoming year was.
The same goes for the tax side of the ledger. Each year, the scorekeepers build an economic forecast and then construct a revenue baseline based on how much the current law tax rates will bring in based on that expected level of economic activity. Proposed changes in tax law are then scored against how much they would reduce or increase revenues versus the baseline.
What about control mechanisms, then?
When it comes to control mechanisms, they are different for mandatory versus discretionary programs. When I was still a teenager, I was an intern for the House Rules Committee in 1990 as the Bush Administration and Congress were finishing a long “budget summit” that was trying to find $500 billion in deficit reduction, much of which would pay for the resolution of the savings and loan crisis.
As part of the settlement, they created two new statutory budget control mechanisms.
For discretionary spending, they wrote into law annual dollar amounts, called caps, that total discretionary budget authority could not exceed each year. There were separate caps for defense, non-defense, and foreign aid programs, so Congress could not shortchange one and beef up another. There was no baseline here, just total spending ceilings each year.
For mandatory spending and taxes, they created a separate control mechanism, called pay-as-you-go budgeting, or PAYGO. Every new law that affected mandatory outlays or revenues would be scored against the baseline, and the aggregate amount of deficit reduction or increase would be written on a 5-year scorecard. The net total of laws enacted by Congress in a given year were not allowed to cause a net deficit increase on the scorecard.
If either of these mechanisms were violated – if Congress enacted discretionary appropriations that exceeded a cap level, or if they enacted mandatory spending or tax laws that added up to a deficit increase versus baseline, then a round of automatic, across-the-board spending cuts, called budget SEQUESTRATION, would be ordered, and the spending cut percentage would be set at the level of the cap or PAYGO violation.
How have these systems worked in practice?
These systems, created in 1990, worked pretty well, to the point that they helped balance the federal budget by 1997. Once the budget was balanced, Congress slacked off on enforcement, and then allowed PAYGO and the spending caps to expire in 2002.
PAYGO was revived permanently in 2010, and the discretionary spending caps were brought back for a decade as part of the resolution to the 2011 debt limit crisis.
As of this recording, there are still no statutory caps on discretionary spending, but it is very possible that the endgame of the coming debt ceiling crisis in 2023 will bring them back.
So what is the weakness of something like PAYGO or discretionary caps?
The weakness of both the discretionary caps and PAYGO is that Congress is free to designate any amount of discretionary funding, or any change to taxes or mandatory spending made in another law, as an “emergency” that is not counted towards cap or PAYGO totals. The only restrictions are that an emergency designation takes 60 votes in the Senate and that the President has to be willing to sign it.
(Incidentally, there are many provisions of budget law that can only be waived in the Senate with 60 or more votes. Simply getting rid of the filibuster in Senate rule 22 will not get rid of these other 60-vote thresholds.)
To recap:
- DISCRETIONARY spending is the kind provided in annual appropriations bills AND where Congress has the discretion to zero out any account or program at any time without exposing Uncle Sam to legal liability.
- MANDATORY spending is where spending levels cannot be changed in the annual appropriations bills unless Congress also enacts changes to benefit formulas or other spending authority in underlying law.
- A BASELINE is an official set of detailed projections of how much money every federal program will spend and how much every federal tax will bring, for the upcoming year and several years into the future (nowadays, it’s usually 10 years). Changes in mandatory spending are judged against a baseline.
Great, thank you so much Jeff. Make sure to tune in next week where we give you an update on the federal budget process during Congress’ lame duck session.