CBO—Still Out of Paradigm after All These Years

By Scott Fullwiler

The Congressional Budget Office (CBO) published its long-term deficit and national debt projections last week.  These are the projections most widely cited in policy discussions about long-term “sustainability” of the national debt and entitlement programs.  In this post I focus on a small but very important part of the report—the CBO’s discussion of the “Consequences of a Large and Growing Debt,” which can be found on pages 13-15.  This section can be found in past reports going back several years, and hasn’t change much if it has changed at all during this time.  It is also consistent with the thinking of most economists on these issues.  As readers of this blog will recognize, the CBO’s analysis is “out of paradigm” in that it is inapplicable to a sovereign, currency-currency issuing government operating under flexible exchange rates such as the US, Japan, Canada, UK, Australia, etc.

CBO presents four consequences of a large and growing national debt.  I discuss each in turn.

  1. Less National Saving and Future Income–Large federal budget deficits over the long term would reduce investment, resulting in lower national income and higher interest rates than would otherwise occur. Increased government borrowing would cause a larger share of the savings potentially available for investment to be used for purchasing government securities, such as Treasury bonds. Those purchases would crowd out investment in capital goods—factories and computers, for example—which makes workers more productive.

This would be laughable if it weren’t for the fact that most economists believe it and the dangers of following policy based on such a belief.  The analysis is based on the loanable funds market—which DOES NOT EXIST in the real world.  In reality, the funds that banks lend are created out of thin air, not constrained by saving, the flow of deposits, or fractional reserve requirements.  Even a 100% requirement changes nothing as long as the central bank is targeting the interbank rate that sets the banks’ cost of funds.  More reserves are always forthcoming via open market operations if the interbank rate starts to move above the central bank’s target, so there is no rise in interest rates of the sort that the loanable funds model supposes.

So there is no threat to funding available for private investment in capital goods, and no threat to the growth rate of future national income.  CBO’s analysis is simply inconsistent with how the modern financial system actually works.  (My post here from 2012 described the process of bank lending. For more detailed analysis of the interaction of banks and central banks, see here, here, here, and here.)

As an aside, consistent with neoclassical theory in which factors of production receive their marginal product, CBO writes that “because wages are determined mainly by workers’ productivity, the reduction in investment would reduce wages as well [since the investment is driving productivity of workers in CBO’s analysis], lessening people’s incentive to work.”  Wow.  Apparently CBO hasn’t noticed that (a) wages and productivity have diverged for the past 40 years, with productivity far outstripping wage growth, and (b) low wages haven’t lessened the incentive to work at all (if they did, US workers should be working far less than there French and German counterparts while the opposite is true).  Three words—Cambridge Capital Controversies, which neoclassicals still haven’t actually bothered to understand.  Instead they hide behind a theory that suggests CEOs making exponentially more than the average worker somehow “deserve” this excessive pay even as anyone that bothers to look can see that these same CEOs have led the private sector to far slower growth rates of productivity than we saw in the 1950s and 1960s when the ratio of CEO pay to worker pay was far smaller.  But I digress.  (And apologies for the rather simplistic analysis here—I realize there’s much more going on with wages—but again, this isn’t the main point of the post.)

  1. Pressure for Larger Tax Increases or Spending Cuts in the FutureWhen the federal debt is large, the government ordinarily must make substantial interest payments to its lenders, and growth in the debt causes those interest payments to increase. (Net interest payments are currently fairly small relative to the size of the economy because interest rates are exceptionally low, but CBO anticipates that those payments will increase considerably as interest rates return to more typical levels.) 

In other words, when interest rates rise, they will take up a larger percentage of the government’s outlays, increasing the likelihood of future large deficits unless spending is cut or taxes are raised.  Similarly, a desire to raise spending or cut taxes in the future will be thwarted by projections of even larger deficits and require still greater cuts or taxes elsewhere.  CBO wants us to believe that it is just trying to protect us from the difficult political decisions this would bring.

In some ways this is a legitimate point, but I would say it differently.  Any increase in the government’s deficit can result in greater aggregate spending on existing productive capacity, so if the government is sending more interest to bond holders, ceteris paribus, this is creating the potential for inflation.  However, the issue here isn’t the larger deficits as much as it is the larger deficits relative to the inflation threat.  But CBO presents us with no analysis of the future inflation threat of rising debt service—in fact, its long-term analysis assumes both an economy at full employment beginning a few years from now and very modest inflation throughout even with the larger deficit and debt service projections.  In other words, the real danger of rising debt service or rising government deficits in general (aside from the obvious potential misallocation of the government’s spending) is assumed away by CBO from the start.

Furthermore, rising debt service for a currency-issuer under flexible exchange rates like the US is a monetary policy variable, as I explained here and here.  If rising debt service is pushing the government’s deficit too high, CBO should explain to us why an inflation-targeting central bank is raising the risk of inflation by raising the government’s debt service.

In any case, an economy reaching the point at which a central bank running a Taylor Rule type of interest rate targeting strategy will raise rates should also be precisely when the government is experiencing fairly rapidly declining primary deficits (the deficit aside from debt service)–which is the case in the US’s history—and probably shouldn’t be entertaining thoughts of increasing deficits at that point if low and stable inflation is a serious policy goal.  In most other cases, the central bank shouldn’t be raising rates and the government should be increasing its deficit.  CBO’s assumption of continuous full employment and low inflation mistakenly abstracts from the fact that the real world economy is always in the midst of some stage of a business cycle.

  1. Reduced Ability to Respond to Domestic and International Problems–When the amount of outstanding debt is relatively small, a government can borrow money to address significant unexpected events—recessions, financial crises, or wars, for example. In contrast, when outstanding debt is large, a government has less flexibility to address financial and economic crises—a very costly circumstance for many countries.  A large amount of debt also can compromise a country’s national security by constraining military spending in times of international crisis or by limiting the country’s ability to prepare for such a crisis.

This one’s pretty amazing—seriously, how can someone actually believe this stuff?  First off, as we know, government’s that issue their own currencies don’t need to borrow back their own money.  Second, even if you do think so, as above, the interest rate on this increase in the national debt is a monetary policy variable, not one that is set by markets.  There is no danger of a currency issuing government not being able to finance its deficits in a time of crisis.  And we know that times of war and financial crisis are in particular the times at which safe, default-risk free government debt is at its lowest rate of interest relative to the debt of non-currency issuers.  Third, such crises are also the points at which the central bank typically has its policy rate—and by extension interest rates on the national debt—set at its lowest.  In fact, it is the private sector that experiences such problems in these times, not the currency-issuing governments—just look back to how private credit markets responded to the global financial crisis of 2008-2009 for the most recent example.

The second part of CBO’s rationale here is even more ridiculous—did they not notice that times of war and financial crisis have been the times of most of the largest increases in the US national debt?  Indeed, the real danger is that in a time of such crisis policy makers will actually believe analysis like CBO’s here.  Thankfully, during WWII they didn’t.  They didn’t listen after September 11, 2001.  And they didn’t listen in 2008 (TARP) or 2009 (Obama stimulus—though the CBO-types did in fact keep the Obama stimulus insufficiently small, not that I was necessarily in favor of many of the spending priorities in the bill).  And in every case of policy makers not listening, interest rates remained low while the government ran the deficits it wanted to run.

  1. Greater Chance of a Fiscal CrisisA large and continuously growing federal debt would have another significant negative consequence: It would increase the likelihood of a fiscal crisis in the United States.  Specifically, there would be a greater risk that investors would become unwilling to finance the government’s borrowing needs unless they were compensated with very high interest rates and, as a result, interest rates on federal debt would rise suddenly and sharply relative to rates of return on other assets. That increase in interest rates would reduce the market value of outstanding government bonds, causing losses for investors and perhaps precipitating a broader financial crisis by creating losses for mutual funds, pension funds, insurance companies, banks, and other holders of government debt—losses that might be large enough to cause some financial institutions to fail.  Unfortunately, there is no way to predict with any confidence whether or when such a fiscal crisis might occur in the United States. In particular, there is no identifiable tipping point in the debt-to-GDP ratio to indicate that a crisis is likely or imminent. All else being equal, however, the larger a government’s debt, the greater the risk of a fiscal crisis.

Here we see the “US could become Greece” argument, with “we can’t say when it could become Greece, but we don’t want to find out!” added on.  In fact, the CBO in the footnotes links to its 2010 report, “Federal Debt and the Risk of a Fiscal Crisis” (in which it makes the same four points as here, by the way, regarding the consequences of large and rising debt), which analyzes recent fiscal crises in Argentina, Ireland, and Greece and then considers how their difficulties dealing with rising interest rates on the national debt, diminished access to financial markets, etc., could harm the US economy.

Again, though, a currency-issuing government under flexible exchange rates can’t have such crises because it doesn’t need to borrow its money; interest rates on its debt are a monetary policy variable.  The doomsayers have been at this for decades now, but have not explained why the US, UK, and Japan ran continually large deficits starting in 2008 at low interest rates while Greece, Spain, Italy, etc., could not.  Their only response is, “Just wait!  This time is NOT different!”  At least CBO doesn’t fall into the typical trap of citing the Reinhart/Rogoff paper on “tipping points,” which has been discredited (see here and here as well).  CBO simply notes here that as of yet “there is no identifiable tipping point”—this is true of course, since there isn’t a tipping point at all if it’s your own currency and you have the ability to set the interest rate on it.  At some point one would think the “US could become Greece” argument would be widely recognized as fraudulent, but if you’re in the wrong paradigm it’s difficult to accept even a simple explanation of why the paradigm is wrong.

In the end, what we see from these four points made by CBO is that the real danger to policymakers isn’t large deficits and debt.  The real danger is that they will pay attention to analysis done of large deficits and debt by CBO and others like it—such as most economists—and unfortunately they are all paying attention and echoing this same sort of analysis.  And here we all sit in a six year trough relative to potential GDP, continued high unemployment (particularly if you include underemployment, etc.) and low participation rates, and with even fairly decent job creation that however is focused on the low-wage end compared to previous recoveries (see here).  And this isn’t even to mention the Eurozone nations that are still in depression states.

22 Responses to CBO—Still Out of Paradigm after All These Years

  1. Erick Borling

    One of the most telling responses indicating the nat’l debt hysteria is balogna is in response to the question “At what debt/GDP ratio do the bond vigilantes appear?” So… Reagan started the nat’l hysteria when the nat’l debt was about two trillion and Oh My God if it gets any bigger we’re all in Serious Trouble! If that’s true, why has the fearsome disaster not occurred? WHEN will that happen, Balanced Budgeteers? As it stands the only trubble we’ve suffered as a consequence of deficit spending (in an attempt to reduce the nat’l debt) has been the irrational and voluntary GOP-created gov’t shutdown. Nothing to do with bond vigilantes or a refusal of people to “lend” to the Federal gubmint. There is no lack of demand for U.S. gov’t debt from domestic or int’l sources, even with incredibly low interest rates. Correct me if I’m rong about any of this.

    • Scott Fullwiler

      I would generally agree. Though I’d say that the problems have been on both sides of the aisle in terms of misunderstanding deficits/debt. Remember Gore’s lockbox? Or how the democrats railed at Bush jr for his tax cuts raising deficits and debt? And Obama’s ready to dismantle entitlements because he thinks we’re out of money.

      • Nat Uerlich

        It’s possible that the fundamental cause of Mr. Obama’s readiness is an erroneous paradigm. But it’s not likely–not in a system where, among other things, public elections are funded privately, most of the money comes from a rich sliver of the population, that tiny minority generally opposes “entitlements” (unless they can be profit centers for the 1%).

        P.S. Progressive thinkers beware: “Entitlements” is a term that has been successfully re-framed to attack Social Security, Medicare, etc.

  2. Scott,

    You’re so polite. The real danger are these idiots. 😉 Isn’t the CBO the same outfit that predicted the Clinton surpluses would continue for 15 years and render the US a financial Shangri-la?

    I have a dumb question to ask you. I have not had time to read the two links you offered that might answer it. When you say ‘rising debt service’, are you talking about the interest owed on treasuries? If so, Frank Newman wrote that the Treasury computes the coming annual interest payments for outstanding treasury securities, then simply issues more treasury securities to meet these obligations. Again, if so, does this amount get added to the deficit?

    • Scott Fullwiler

      Sounds like Frank is talking about refinancing tsys that are coming due, not debt service (which, yes, is interest on tsys).

      • Hi Scott,

        I just looked it up, and actually answered my own question about whether it adds to the deficit (it does and it doesn’t depending on what you’re looking at) . It’s in Chapter 6 of his book: Freedom From National Debt. The chapter is titled “Why interest on Treasuries is not a problem for the U.S.” (p. 47)

        He writes (I highlighted this, but he explains it in far more detail in the chapter):

        If the economy is running with substantial unemployment, then it is better for the government to issue more Treasuries to pay for its needs rather than raise taxes. That includes paying interest on existing Treasuries by issuing new Treasuries. There is no more or less money in the system, on a net basis, and no need for additional taxes.

  3. “If rising debt service is pushing the government’s deficit too high, CBO should explain to us why an inflation-targeting central bank is raising the risk of inflation by raising the government’s debt service.”

    This is a really fascinating conclusion. It is a fairly strong argument in favour of ZIRP forever. But, doesn’t this formulation ignore the impact of CB rate-setting on private sector credit creation? The higher CB rate increases inflation by increasing gov’t deficits, but the higher rate is used by banks as a baseline for rates to their own customers. Taken far enough (as in 1982), that would drive those customers into bankruptcy, launch an asset fire-sale, and so would be deflationary. Yet another argument in favour of ZIRP forever.

    • Isn’t the relevant number for monetary inflation not the combined rate of growth of both public and private sector money creation rather than just one or the other?

      • Scott Fullwiler

        Generally, yes. Spending from bank lending is still spending just as much as spending out of the income from government interest payments is.

        I don’t know that there’s a particular formula setting both types of money to inflation, though I’ve said before that MV = PY has the wrong M and V. The M is the govt’s deficit and V is an inverse function of the non-govt sector’s desire to net save.

        However, one could say that one is govt money and the other is bank money. In that case, might call it H (for horizontal bank money) and V (f0r govt vertical money) and then say HV = PY. But it’s not a direct relationship, so it would be f(H,V) = PY, and H would be an inverse function of pvt sector’s desired net saving.

        Hope I didn’t confuse you there–just kind of thinking out loud.

        • The formulation of M does seem to be a matter of some dispute, which I think is what my question was focused on — the broadness of that definition. But as you write, one could define V in a manner that allows a less broad definition of M. Now I’m stuck wondering if the long-term decline in personal savings rate, which may be inversely changing V, is relevant in describing change in P, and excessive focus on M as the driver of P is misplaced.

        • Changes in monetary stocks cannot explain changes in spending. To illustrate, what changed the united states economy between 2007 and 2009 was not the stocks of money, but the value of the houses. All asset classes play part in determining aggregate demand.

          And, as investing increases net wealth by creating tangible assets, it is not far stretch to theorize that raise in net wealth of the private sector is the ‘transmission mechanism’ that raises spending in both deficit spending and bank lending. The stock of housing debt for example is quite obviously grown alongside value of the houses that the debt has financed, value of which partially neutralizes the need borrowers need to pay back their loans as their net wealth position stays acceptably high even with debt.

          And out of this process financial assets flow, like from fiscal stimulus, to the rest of the economy. Maybe it is the most important flow of the economy, flow of assets, from where they are created, to where they are saved.

  4. Regarding point 1, I want to make the argument that the first sentence –

    “Large federal budget deficits over the long term would reduce investment, resulting in lower national income and higher interest rates than would otherwise occur”

    does not rely on the loanable funds view you critique and so requires further argumentation to be invalidated. I would say that the first sentence stands independently of the second and third sentences, and so even if the second and third sentences are incorrect (and I’d agree with you that they are, for the reasons you lay out), that is not sufficient to disprove the first.

    Start from the national income accounting identity (assume closed economy for simplicity): Y = C + I + G.

    Call aggregate supply Y and aggregate demand: C_d + I_d + G (C_d and I_d are desired consumption and investment). The goods market is in equilibrium when aggregate supply equals aggregate demand, otherwise we experience inflation or deflation. I wouldn’t expect this to be overly controversial with MMT, even if you prefer a different model – it’s the same as saying we get inflation/deflation (perhaps due to excessive/insufficient government spending) if aggregate demand exceeds/is less than aggregate supply.

    The alternative form of this is that desired national saving has to equal desired investment. We do some simple algebra to transform the equilibrium condition [Y = C_d + I_d + G] to [Y – C_d – G = I_d]. The left hand side is the same as desired national saving: S_d = I_d.

    We can then plot the S_d and I_d curves with the real interest rate on the y axis. According to Bernanke’s textbook, the empirical evidence is controversial, but the most widely-accepted conclusion is that an increase in real interest rate reduces current consumption and increases saving, but that the effect isn’t very strong. So S slopes up. I slopes down with the real interest rate if you think the user cost of capital is a reasonable determinant of desired investment. I could see someone disagreeing with these relationships, but you’d need to make a strong empirical argument at the end of the day explained by a robust theory.

    So now suppose Y is fixed (perhaps we’re at full employment output), and the government spends. Desired national saving S_D = Y – C_d – G is therefore lowered, for any prevailing real interest rate. Graphically, this is a shift upward in the saving curve. We presume desired investment does not change if the expected real rate does not change.

    Therefore, we’re in a situation where aggregate demand exceeds aggregate supply at the current real interest rate. Desired investment exceeds desired saving. In other words, Y is fixed, desired investment hasn’t changed, and aggregate demand (C_d + I_d + G) has only increased.

    The real rate would have to increase to bring the goods market back into equilibrium. This is where most textbooks bring loanable funds into play (Bernanke et al doesn’t even use this term, but the explanation is similar). The problem with it is exactly what you say: in our modern financial system, there is no natural market mechanism related to S and I (as theorized in the loanable funds model) by which the real rate would adjust if the goods market is out of equilibrium. Instead, what has to happen is that the central bank has to increase their target rate to make it so. But an increase in the target rate will lead to lower investment in the future, if you agree with the user cost of capital argument above.

    So the statement

    “Large federal budget deficits over the long term would reduce investment, resulting in lower national income and higher interest rates than would otherwise occur.”

    still stands and does not rely on the flawed logic of the loanable funds interpretation that you critique. As far as I can see, if you find the above model somewhat reasonable, it would seem the angle at which you could challenge this premise would have to do with the shapes of the S and I curves or the long-term flexibility of Y. If you find the model too flawed to be useful, then you’d have to make a different argument.

  5. Scott Fullwiler

    Hi ATR,

    Thanks for the comment. I’m going to respond in parts:

    “does not rely on the loanable funds view you critique and so requires further argumentation to be invalidated. I would say that the first sentence stands independently of the second and third sentences, and so even if the second and third sentences are incorrect (and I’d agree with you that they are, for the reasons you lay out), that is not sufficient to disprove the first.”

    OK from your perspective or for the sake of argument. But if you read the entire text of the CBO report on this, they are clearly referring to govt crowding out investment by sucking up saving. The first sentence is in their thesis whereas the others are the logic that supports it, in CBO’s view. Again, though, for the sake of argument, I’ll go along–it could certainly be that you COULD argue the first sentence without the others, which is obviously what you are doing here. that’s fine, but I don’t think that’s what CBO is doing.

    “Start from the national income accounting identity (assume closed economy for simplicity): Y = C + I + G.”

    OK

    “Call aggregate supply Y and aggregate demand: C_d + I_d + G (C_d and I_d are desired consumption and investment). The goods market is in equilibrium when aggregate supply equals aggregate demand, otherwise we experience inflation or deflation. I wouldn’t expect this to be overly controversial with MMT, even if you prefer a different model – it’s the same as saying we get inflation/deflation (perhaps due to excessive/insufficient government spending) if aggregate demand exceeds/is less than aggregate supply.”

    OK.

    “The alternative form of this is that desired national saving has to equal desired investment. We do some simple algebra to transform the equilibrium condition [Y = C_d + I_d + G] to [Y – C_d – G = I_d]. The left hand side is the same as desired national saving: S_d = I_d.”

    OK

    “We can then plot the S_d and I_d curves with the real interest rate on the y axis. According to Bernanke’s textbook, the empirical evidence is controversial, but the most widely-accepted conclusion is that an increase in real interest rate reduces current consumption and increases saving, but that the effect isn’t very strong. So S slopes up. I slopes down with the real interest rate if you think the user cost of capital is a reasonable determinant of desired investment. I could see someone disagreeing with these relationships, but you’d need to make a strong empirical argument at the end of the day explained by a robust theory.”

    Here’s where I don’t agree and it is right in line with my critique above. You’ve assumed that S determines I or “creates the funds” for I, and/or that S and I together determine the interest rate. I’ve argued throughout this post that none of that is actually true. I’ve also argued it in the various links I’ve provided. There’s also 60 years of Post Keynesian research on this, so I think the “theory” is robust enough, but you really only need to understand how the balance sheets actually work to know it’s true and a model setting interest rates with I and S is wrong. Also, no I don’t think the user cost of capital is a reasonable determinant of desired investment–again, capital controversies and/or Fazzari’s research.

    “So now suppose Y is fixed (perhaps we’re at full employment output), and the government spends. Desired national saving S_D = Y – C_d – G is therefore lowered, for any prevailing real interest rate. Graphically, this is a shift upward in the saving curve. We presume desired investment does not change if the expected real rate does not change. Therefore, we’re in a situation where aggregate demand exceeds aggregate supply at the current real interest rate. Desired investment exceeds desired saving. In other words, Y is fixed, desired investment hasn’t changed, and aggregate demand (C_d + I_d + G) has only increased.”

    I would be fine with an argument that suggests the government spending uses real resources relative to a fixed real capacity to produce goods and services. That’s what I said regarding CBO’s point 2–there could be inflation if the govt deficit is too big to push AD too far relative to AS. But you are mixing the financial crowding out argument with the real goods and services crowding out argument. The latter can most definitely happen; the former cannot.

    “The real rate would have to increase to bring the goods market back into equilibrium. This is where most textbooks bring loanable funds into play (Bernanke et al doesn’t even use this term, but the explanation is similar). The problem with it is exactly what you say: in our modern financial system, there is no natural market mechanism related to S and I (as theorized in the loanable funds model) by which the real rate would adjust if the goods market is out of equilibrium. Instead, what has to happen is that the central bank has to increase their target rate to make it so. But an increase in the target rate will lead to lower investment in the future, if you agree with the user cost of capital argument above.”

    I would say that you’d have to slow the economy down because AD is too high relative to AS. Again, that’s fine. I don’t think the real rate is the way to do it, as I noted above, but within a neoclassical model, yes, that’s how they do it. Again, I think that’s CBO’s argument in their second point, not this one–and if I’m wrong, then they’ve confused financial crowding out with real crowding out, which is what neoclassicals do all the time, so no surprise there.

    “So the statement “Large federal budget deficits over the long term would reduce investment, resulting in lower national income and higher interest rates than would otherwise occur.” still stands and does not rely on the flawed logic of the loanable funds interpretation that you critique. As far as I can see, if you find the above model somewhat reasonable, it would seem the angle at which you could challenge this premise would have to do with the shapes of the S and I curves or the long-term flexibility of Y. If you find the model too flawed to be useful, then you’d have to make a different argument.”

    Again, I went at the real crowding out argument in relation to CBO’s second point. I think it’s a valid one potentially, but I also think they’ve screwed up by assuming full employment forever rather than business cycles. And I don’t think S and I setting r is a good model at all, because it doesn’t work that way.

    Now, if you want to say that there’s an r that will give us the right amount of I and through that S (i.e., I causes S instead of I being constrained by S) to get to full employment, then that’s better and at least then you’re not making the financial crowding out argument (which perhaps that is what you’re saying throughout here, and if so I apologize for misinterpreting above). But even then I’m not going to worry about shapes of I curves and S curves because I don’t think there’s an r that gets us to the right level of I and S. Not to mention that I don’t know which r we’d be talking about (overnight rate? 10y treasury? AAA corporate bond rate? mortgage rate?). And the cost of capital is far less important than expected free cash flows for investment–otherwise we should have seen massive I the past 5+ years, and we obviously didn’t. To clarify, I think yOu certainly do need macro policy to get you to the right AD, but I think it’s more complicated than setting an r

    • Scott Fullwiler

      HI again, ATR

      Sorry about continuing here, but I may have been misinterpreting your comment throughout. So, I’ll try to clarify . . .

      If you’re simply saying that CBO’s arguing that the deficit pushes the economy too far and thus r has to rise to pull it back (I’m leaving out the S and I curves stuff for the moment), then that’s fine if we’re talkinb about the CB raising r in response (again, I don’t think raising r will do it very efficiently, but let’s ignore that here).

      But I do think this is more the second point they are getting at (and the 3rd, actually, too, if one reads beyond the first paragraph that I’ve copied above). In the 2nd sentence of the first point, they say that “Increased government borrowing would cause a larger share of the savings potentially available for investment to be used for purchasing government securities.” As I noted, the first sentence is their thesis, and the rest are how they get there (and I’ve only pasted the first paragraph of their discussion for each). And the second sentence is pure loanable funds, obviously.

      Yes, as above, I do think you could make the same argument from within the neoclassical paradigm for S, I, and r as you do without invoking loanable funds or S constraining I (which is how I am now interpreting your comment–hope I’m close to right this time!), but you have to invoke r as a policy variable then to do that (i.e., the CB needs to find the right r, and the right r is now higher b/c of the deficit). And CBO is definitely not doing that as point 4 makes clear; if r is a policy variable, then point 4 doesn’t matter, and they obviously think it does.

      OK, I’ll just wait for your reply in case I’ve misinterpreted for the second time instead of clarifying.

      • Yup – you’ve got it in your second comment. I’m not saying the authors aren’t making errors. Rather, I’m trying to lay out how they *should have* argued from within their own neoclassical model. I agree that in context they use the 2nd and 3rd sentence to support the 1st sentence, and I also agree sentences 2 and 3 are wrong on their face. I’m just saying the first sentence can stand without them within the neoclassical model, and you can use the S/I model to show why without invoking the typical loanable funds ideas. You can probably (?) find neoclassical economists who understand your points about the financial system but nonetheless still believe in such a model. More on that in a follow-up comment that I’d be curious to get your thoughts on.

        So all this to say is that the S/I model is not as terrible as the typical loanable funds explanation makes it seem, even if you still don’t like it for other reasons (but you have to have other reasons). We can throw that stuff out the window and give it a second chance for a more coherent argument, which would be along the lines of: “Desired S and desired I are in part determined by interest rates, and interest rates are determined by the central bank. There is a certain interest rate at which desired S and desired I would be equal (which is the same as saying aggregate demand would equal aggregate supply). It’s the central bank’s job to find that rate. ”

        No doubt neoclassical economists should understand the rebuttal you make, but if they were worth their salt, they’d concede on that point but still challenge you to argue why the interest rate the central bank sets is not an effective tool to equilibriate desired S and desired I. And perhaps they so often don’t give your anti-loanable funds argument a chance because what they’re really latching onto is the above, even if they don’t have 100% intellectual clarity on the distinction. Just my current working hypothesis, though.

        • My follow-up comment, but it’s a total side-note:

          So Nick Rowe has a story of loanable funds that doesn’t have to rely on a flawed understanding of bank lending. It does rely on a barter economy, which I realize you’d object to, but for the sake of dispelling the financial loanable funds view:

          Pretend there’s a barter economy, and lending/borrowing is done in real goods instead of money (e.g., if the real rate of interest was 2%, you’d pay the loan back by providing 2% more real goods in the future). So for example, say agg demand exceeded agg supply. You’d then have a bunch of people looking to borrow real goods currently, to satisfy the excess demand. The real rate of interest on the goods would be bid up, until agg demand = agg supply. So there you have a natural market mechanism that adjusts the real rate to achieve goods market equilibrium.

          Why you’d choose to call that “loanable funds,” I don’t know, but that’s what loanable funds is in a barter economy. It’s a more coherent story than the typical financial loanable funds story told, setting aside objections to the usefulness of a barter model.

          But that made me think – maybe the financial loanable funds story we see in textbooks is an extremely short-handed but flawed attempt to translate what would happen in a barter economy to what happens in a financial economy. Perhaps based on a flawed understanding of banking, authors thought they could get away with the ‘saving funding investment’ analogy, and drop the barter story altogether. But all that is is a bad analogy – not a fundamental flaw in the model itself, properly understood. Instead, what these authors should do is tell the story of how rates adjust in the pretend barter economy, but then make clear that it doesn’t translate to the modern banking system. Instead, banking would be shown to work as you describe, and it would be argued that it’s the central bank’s job to change rates as would naturally occur in the barter economy.

          I’m not arguing the above model is right. It’s just my desperate attempt to reconcile it with banking operations as MMT sees it.

          • Reading your first reply to me, it would seem you might agree to some extent. This is my hypothesis for how the loanable funds story came to confuse real crowding out with financial crowding out.

      • Jerry Brown

        To clarify my muddledness, if you have a fiat money system which allows banks and the government to create money at their discretion, can’t they always create money for investment even if there isn’t money saved beforehand?

    • Jerry Brown

      “The alternative form of this is that desired national saving has to equal desired investment” then some algebra, then you say “OK”. Why ok? Wouldn’t that only be the case if you assume that the economy was already running at full capacity without any unusual inflation? Even then, would it maybe be better to say that national savings Will equal investment? Or is it ok to say that saving has to equal investment , which implies that investment is always dependent on savings, at least to me. Please help me out with this one, I have gotten my thinking all muddled up with this.

  6. Pingback: CBO – Still Pushing Deficit Scaremongering Propaganda | naked capitalism

  7. Justin Santopietro

    Hi Scott,

    Don’t know if you saw this, but I tried to take down CBO in this satirical post I did a while back:
    http://www.dailykos.com/story/2014/07/13/1313771/-BREAKING-NEWS-CBO-warns-of-impending-email-scarcity

    With lunacy of this high degree, I think humor is the only adequate response. As someone who has seen the legislative process up close, its disgusting to me how much power CBO wields on the Hill. People on both sides of the aisle have to wait on baited breath for a CBO ‘score’ of their bills, and then take their worthless macro projections as gospel. Its the epicenter of institutionalized stupidity in the US. Thanks for taking another swipe at them!

  8. Scott, I wanted to ask what you meant when you said

    “It’s always important to keep in mind that traditional monetary policy “works” by encouraging more/less spending out of existing income, whereas fiscal policy works first by adjusting income, then spending through that.”

    Because spending equals income something is amiss here.