I get many E-mails every week from people asking me to explain exactly what a deficit is. They understand that a budget deficit is the difference between revenue and spending but then become confused as a result of being so ingrained with narratives emanating from politicians and lobbyists who misuse terms and always try to conflate deficits and debt. So today’s blog is a basic primer on deficits and why you should welcome them (usually) and why we all should sleep tight when the government is in deficit. So – budget deficit basics …
There were two recent articles in the Wall Street Journal that motivated me to write this blog today rather than another time. They were examples of political statements which completely abused the underlying economic concepts being discussed and therefore mis-informed the public debate.
In doing so, they self-reinforced illegitimate economic concepts and further placed the respective economies being focused on in further jeopardy.
The first (April 1, 2011) – Japan Government Slams Calls For BOJ Underwriting Of Public Debt – reported statements by the top-ranked ministers in the Japanese government in relation to how the resources need to pursue the earthquake and tsunami reconstruction effort would be garnered (paid for!).
The opposition Democratic Party of Japan has signalled that the “concerns over the impact of BOJ underwriting are overblown”. The DPJ want the Bank of Japan (central bank) to “print more money to underwrite government bonds–a step that would require approval by the parliament”.
I recently considered the reconstruction issue in this blog – Earthquake lies and then specifically, the issue of the Bank of Japan directly buying Ministry of Finance debt in this blog – We’re sticking to our strict fiscal rules.
The WSJ article first quoted the Finance Minister Yoshihiko Noda who said:
The government isn’t considering … [any direct central bank financing of public debt official must be] … very, very careful going forward …
The WSJ also said Noda noted that “long-term interest rates rose as soon as a report that lawmakers were considering direct BOJ bond purchases was published electronically Thursday”.
The reference to long-term interest rates is an attempt to say that the “markets” fear inflation would result from the direct purchase of public debt by the Bank of Japan. Long-term interest rates depart from short-term rates by some risk component and long-term inflationary expectations.
Rising long-term yields on public debt may also signal an improving economy which leads bond investors to diversify their portfolios into riskier financial assets and so the demand for public debt drops and yields rise.
Long-term bond yields are at historic lows at present and small upward corrections in yields do not mean that the bond markets are fearing an inflation outbreak any time soon.
The yield on 10-year notes (currently at 1.3 per cent) has already crept up by nearly 10 basis points over the last 12 months but remains well below the average over the last 25 years (around 2.75 per cent). Just before the property crash in 1990, the bond yield for 10-Year notes was 8.23 per cent. As the deficits expanded in the late 1990s to restore growth the yields plunged and by June 2003 had reached a record low of 0.45 per cent.
The WSJ also reported that the Economy Minister Kaoru Yosano said that “having the central bank underwrite government securities” would be:
… unthinkable, and I won’t let it happen, no matter what … such action, ignoring fiscal discipline, the Japanese government and the BOJ would lose international credibility … I can’t even forecast how that would ricochet against long-term interest rates in the market.
The question arises that why would international credibility matter (about what anyway?) when Japan places almost all of its debt in the hands of domestic investors who cannot get enough of it?
But these ministerial statements are merely political statements reflecting the hold that neo-liberals have on the policy debate. There is no credible economic content to the statements.
I note that today – the first trading day after the Ministers made their statements about re-affirming their commitment to “fiscal discpline” (conservative-style) that the – Japanese government yield curve – has flattened a bit (short-term yields down and long-term yields up).
If these movements last week really reflected the fear that inflation was about to become a problem in Japan as a result of speculation that the central bank might start directly buying Ministry of Finance debt, then the yields should have fallen this week as a result of the strong statements from the two senior economic ministers.
The point is that these movements have nothing to do with the speculation that the BOJ might exercise its legal right and directly facilitate Japanese government spending without forcing it to match the deficit by debt-issuance into the private bond markets.
The misunderstandings were not only on the conservative side of Japanese politics. The WSJ article quoted a senior DPJ Upper House politician Yoichi Kaneko who said:
The only negative aspect of BOJ underwriting is that prices will rise because of money being spent by the government flowing into the market … I can’t imagine that long-term rates will climb …
So the “progressive” is a quantity theorist – the ultimate deception mounted by conservative economists. If public spending caused inflation as a matter of consequence then Japan should have hyperinflation by now. Instead it has been battling deflation for years not as the budget deficit have been maintained.
The article rounded off with hints that the suggestion that the Bank of Japan purchase the public debt has arisen because of insolvency risks:
The concern is that a further increase in Japan’s accumulated public debt–already twice its annual economic output–will likely fuel concerns over the government’s ability to stem its borrowing binge, prompting investors to demand more interest payments to compensate for the risk of holding government securities.
Who is expressing concern here? The bond markets seem content. The people in employment that is being supported by the on-going government deficits are benefiting. The regions that will be reconstructed will benefit.
I just love journalists who cite “the concern” but don’t tell you who is concerned other than a group of over-paid financial market types who want to drive yields on their asset holdings up. The bond markets know that their purchases of Japanese government debt is risk-free – otherwise they would have stopped buying it decades ago.
Historically, long-term bond yields (as noted above) have been very low relative to elsewhere.
But the point is that the article is perpetuating the insolvency risk myth. In one short article, we see the major myths of mainstream macroeconomics in terms of budgets being rehearsed. A solid concentrated dose of lies.
When this sort of nonsense is being peddled every day by a multitude of media commentators it is no wonder that people get confused. Especially when those on the progressive side of politics pump out their own version of the same narrative.
And then you go to a dinner party on a Friday night … and someone there has studied economics somewhere and they spout forth the “wisdom” they rote-learned from Mankiw’s textbook and you spend the rest of the weekend writing E-mails to me asking for some oblation! It is no wonder.
The second (April , 2011) – GOP Aim: Cut $4 Trillion Budget Plan Would Transform Medicare, Reset Budget Debate; Democrats Balk – discussed the 2012 budget proposal by the US Republicans which “would cut more than $4 trillion from federal spending projected over the next decade and transform the Medicare health program for the elderly”.
I don’t intend to critique that article – the points are obvious.
They plan to deny the poor and weak of real resources that are necessary to deliver a modicum of respectability because they claim they cannot afford them. Well they claim that providing this semblance of dignity and security to citizens who cannot otherwise provide it themselves would be “too costly”.
They do not actually understand that the “cost” of these programs is the real resources that are consumed by them. The dollar numbers attached to the programs are meaningless in fact. The US government can always “afford” to “spend” the required sums.
But you will never hear the debate constructed in terms of whether these real resource allocation are possible or not. It is possible that the growing real resource requirements will require political decisions as to whether the US community is getting the best value out of those resources (in that particular use).
That is a legitimate debate but has nothing to do with budget deficits.
Neither side of politics in the US is pointing that out.
So what exactly is a budget deficit?
It is the net outcome of two broad flows (denominated in the currency that the national government issues) – revenue and spending. What do we mean by a flow? Many economic graduates fail to differentiate between a stock and a flow.
A flow is a variable that is measured per unit of time. So a certain number of dollars being spending per week, month, quarter. If a government was spending $A10 million per day then after a week they would have spent $A70 million and so on. You add flows together to get a lower frequency aggregate (a week is a lower time frequency than a day).
Once a flow has occurred it has gone. It has consequences but a flow is not “paid back”.
Alternatively, a stock is a variable that is measured at a point in time. So a dollar amount at the end of 2010. If there are 1 million people unemployed in January, 2 million in February and 3 million in March – these are stock measures. To get a measure of unemployment in the first quarter of that year (January-March) we would average the individual stock observations to get a value of 2 million.
A stock is a legacy. Flows create stocks and a well-grounded macroeconomic framework has to be stock-flow consistent. Please read my blog – Stock-flow consistent macro models – for more discussion on this point.
So job hires (the number of workers entering a job per period) and quits (the number of workers leaving a job per period) lead to stock measures of employment and unemployment (once we know what the labour force is.
Similarly, depending on specific arrangements in place, a rising budget deficit (the net sum of the revenue and spending flows) can add to the stock of outstanding public debt (a stock) if the government insists of issuing debt $-for-$ to match the net public spending (the deficit). If the government didn’t have these arrangements in place then there would be no particular relationship between deficits and outstanding public debt.
Commentators get hopefully confused when discussing the economic consequences of the flows (deficits) and the stock (outstanding debt).
But when we are thinking about deficits we should ignore the stock associations. From the perspective of Modern Monetary Theory (MMT) the debt implications are an entirely separate discussion which should not be conducted at the same time as we seek to analyse the consequences of the flows.
So when you hear someone say the deficit is too big (whether in absolute terms or relative terms (scaled to GDP)) you should ask them in relation to what
When you read a politician say that “prices will rise because of money being spent by the government flowing into the market” if the central bank bought treasury debt, you should ask two questions: (a) Does that mean that all spending is inflationary? and (b) What difference does the issuance of debt to private bond holders make relative to no private debt placement?
The answers you will typically get are:
1. Yes, deficits are inflationary – which isn’t an answer to the question asked and tells you that the person doesn’t understand the point.
2. Central banks buying debt is just printing money and so there will be “too much money chasing too few goods” and that will be inflationary. But if you issue debt you soak up that private spending capacity and replace it with public spending and that is less inflationary. If you think about that within the flawed logic you might wonder whether government spending would add anything to demand.
A mainstream economist however will approach the matter in a more analytical manner. I recently wrote (again) about the government budget constraint literature (GBC)in this blog – How are the laboratory rats going? – and so I only state the conclusions here without explanation. You are encouraged to refresh your memory by re-reading that blog.
The mainstream think that the government has to “finance” its spending just like a household. The government is just like a household but bigger from the perspective of a mainstream economist – a super-household.
So they then move onto analysing the “financing” choices of government and integrate these options in their analysis of the consequences of government spending.
To restate, the GBC equation is written as:
which you can read in English as saying that Budget deficit in year t (BDt) equals Government spending (G) + Government interest payments (rBt) – Tax revenue (T) which in turn must be “financed” by the change in public debt (B for bonds) and/or a change in high powered money (H). Here r is the interest rate on outstanding public debt, and Bt is the stock of bonds outstanding at the end of the period in question.
The triangle sign (delta) is just shorthand for the change in a variable and the little t next to the variables refers to time period t (that is, now).
Note that Bt is a stock but rBt is a flow because if, say, r = 0.05 and Bt was $A100 million then rBt would be a flow of government spending equal to 5 million in that particular period (t).
The change in high-powered money would arise if the option that is being discussed in Japan at present – the Bank of Japan directly buying Ministry of Finance debt – was used.
The change in bonds would arise if the government sold bonds $-for-$ (yen-for-yen) into the private market to match the excess of government spending plus interest payments over tax revenue.
The important point is that mainstream macroeconomic theory considers that the consequences of the net public spending differ according to whether the spending is associated (funded in their parlance) by ΔB or ΔH.
Modern Monetary Theory (MMT) disputes that claim.
In terms of stock-flow consistent macroeconomics, the equation above will always hold and in saying that I am not ascribing any particular significance to that result. As long as all the transactions between the government and non-government sector have been correctly added and subtracted then the budget deficit described above will always be reflected by the sum of the changes in ΔM and ΔB.
That is, in terms of MMT, the GBC equation is just an accounting identity that has to be true by definition once all the transactions are accounted for (that is, ex post or after the fact) and has no real economic importance. The spending has significance but the monetary operations (the ΔB and ΔH) do not.
But for the mainstream economist, the equation represents an ex ante (before the fact) financial constraint that the government is bound by. MMT proponents note that this cannot be true as a matter of intrinsic capability if governments issue their own fiat currency. It is clear from my blog last week – We’re sticking to our strict fiscal rules – that the Bank of Japan can facilitate (term used in place of “fund”) government spending directly and the private sector doesn’t have to buy any debt.
Yes, ΔH will rise as an accounting reality but given the Bank of Japan is really part of the consolidated government sector it is nonsensical to consider that as a “financial constraint”. Please read my blog – The consolidated government – treasury and central bank – for more discussion on this point.
It is true that governments might place voluntary constraints on themselves for political reasons – as in the statement made by the Japanese Economic Minister noted above. There are many ideological constraints – justified by the call for imposing fiscal discipline – that governments use to make the operational realities of the fiat currency system more elongated.
But they can be dismantled in the same way they were imposed – by discretionary choice. They are not financial constraints in the way the mainstream textbooks suggest. Please read my blog – On voluntary constraints that undermine public purpose – for more discussion on this point.
From another perspective – the justification used for these voluntary constraints is rather wan to say the least.
The statement by the Japanese Economic Minister – “such action, ignoring fiscal discipline” – is representative of the arguments that have been made following the collapse of the Bretton Woods system in 1971 to justify continued debt-issuance to the private sector – when in operational terms there is no reason to issue such debt.
Under the fixed exchange rate, convertible currency system that ruled up to 1971 (in the Post World War II period) governments were financially constrained. After 1971, most government shed those shackles in operational terms but were pressured by conservatives to continue limiting their fiscal latitude.
But think about it for a moment. These politicians are seeking out vote and then telling us they cannot be trusted to exercise the power we wrest in them. It has always been a nonsensical contradiction.
Anyway, the mainstream economists debate the choice between ΔH (change in high powered money) and ΔB (change in bond issuance) in this way.
The standard framework compares public spending backed by:
- Tax revenue (that is, balanced budget spending).
- Bond issuance (that is, ΔB).
- High powered monetary growth (that is, ΔH).
They claim that the degree of expansion is greater with the third option relative to ΔB, which, in turn, is more expansionary than no deficit at all.
The result that ΔH is most expansionary is based on several flawed assumptions and understandings. They claim that the central bank controls the money supply – because there is a money multiplier operating that links changes in the monetary base (ΔH) to changes in the broader money supply.
They claim that interest rates are determined by the private demand for money relative to the supply of money and under usual conditions when the government expands high powered money, interest rates fall which stimulates private spending in addition to the government spending.
They do claim that the government spending increases the demand for money and pushes interest rates up but the monetary accommodation (ΔH) offsets this to some extent. The debate then is about the extent of these effects.
They compare that with debt-issuance which they say leaves the money supply unchanged and so the only monetary effects are an increasing interest rate (via the government spending pushing up interest rates) which chokes off private spending. This is the crowding out effect. The public spending crowds out private spending.
The extent to which this happens is debated at length (this is what mainstream academics write endlessly about) with moderate mainstreamers allowing government spending to be expansionary but less so than the monetary expansion because of the crowding out effects. The extreme mainstream economists believe the crowding out is 100 per cent. Ricardian Equivalence is a version of that extreme view.
The mainstreamers then say that as high powered monetary creation is the most expansionary it should be eschewed because it will be inflationary. They invoke the Quantity Theory of Money here – which claims there is a direct and proportional relationship between monetary growth and inflation. They forget to remind everyone that the result could only hold if there is full employment (and some other assumptions that also rarely hold).
The problem is that the myth has become part of the folklore that we operate in and so you even hear progressives mouthing that concern (as noted above in the Japan example).
As an aside, by recognising that central banks can directly facilitate government spending without recourse to the private debt markets the mainstream are really admitting there is no financial constraint on government spending. That is why they are so vociferous about inflation.
There are so many errors in logic and fact in the mainstream argument – among the key problems:
1. Central banks cannot control the money supply. Please read my blog – Understanding central bank operations – for more discussion on this point.
2. There is no money multiplier – please read my blog – Money multiplier and other myths – for more discussion on this point.Second,
3. Expanding high powered money does not cause inflation. Please read the following blogs – Building bank reserves will not expand credit and Building bank reserves is not inflationary – for further discussion.
4. The central bank sets the interest rate – please read my blog – Will we really pay higher interest rates? – for more discussion on this point.
5. Increased government spending does not put upward pressure on interest rates – please read my blog – Will we really pay higher interest rates? – for more discussion on this point.
6. The national government does not have a financial constraint and the monetary operations that might be associated with government spending does not alter the impact of that spending.
The government spends the same way irrespective of these operations. It credits bank accounts (or indirectly credits them by issuing cheques). This spending adds to aggregate demand and boosts employment and output – as long as their is increased capacity.
The reality might be that the treasury may instruct the central bank to credit some bank accounts and some intra-government accounting record would be altered. The accounting is of no interest to us economists in this context.
Alternatively, the treasury might instruct some other government agency to put out a bond tender to match the spending increase. The spending still occurs in the same way and then the accounting for the debt issuance is a bit different (as we will see) but of no significant to that spending.
The mainstream claim that high-powered money creation (depicted as the government asking the central bank to buy treasury bonds which the central bank in return then prints money) means that the extra spending will cause accelerating inflation because there will be “too much money chasing too few goods”.
But in fact the economies will respond to aggregate spending increases by expanding output unless they cannot do that. When might that be? At full employment. In which case, why would a government continue to expand nominal spending if they know they cannot tease any more output (and employment) out of the real economy?
Most economies are typically constrained by deficient demand (defined as demand below the full employment level) and so nominal demand increases (growth in spending) will not lead to inflation.
So when governments are expanding deficits to offset a collapse in private spending, there is plenty of spare capacity available to ensure output rather than inflation increases. You will also appreciate that the inflation risk comes from the spending – not what accounting gymnastics are performed – that is, whether the government “borrows from itself” (exchanges of accounting information between treasury and the central bank) or borrows from the public (swapping a bank reserve for a bond account).
So these monetary operations (high powered money creation/debt issuance) neither increase or decrease the risk of inflation associated with the spending. What matters is whether there is spare capacity in the economy to increase production when a spending increase enters the economy. If there is then the inflation risk is low to non-existent.
If there isn’t then the inflation risk is high – bond issuance or not!
But what about the argument that bond issuance is less expansionary because it soaks up private spending capacity?
What would happen if a sovereign, currency-issuing government (with a flexible exchange rate) ran a budget deficit without issuing debt?
Like all government spending, the Treasury would credit the reserve accounts held by the commercial bank at the central bank. The commercial bank in question would be where the target of the spending had an account. So the commercial bank’s assets rise and its liabilities also increase because a deposit would be made.
The transactions are clear: The commercial bank’s assets rise and its liabilities also increase because a new deposit has been made. Further, the target of the fiscal initiative enjoys increased assets (bank deposit) and net worth (a liability/equity entry on their balance sheet). Taxation does the opposite and so a deficit (spending greater than taxation) means that reserves increase and private net worth increases.
This means that there are likely to be excess reserves in the “cash system” which then raises issues for the central bank about its liquidity management. The aim of the central bank is to “hit” a target interest rate and so it has to ensure that competitive forces in the interbank market do not compromise that target.
When there are excess reserves there is downward pressure on the overnight interest rate (as banks scurry to seek interest-earning opportunities), the central bank then has to sell government bonds to the banks to soak the excess up and maintain liquidity at a level consistent with the target. Some central banks offer a return on overnight reserves which reduces the need to sell debt as a liquidity management operation.
These debt sales do not “finance” government net spending – they are a monetary operation aimed at interest-rate maintenance. So M1 (deposits in the non-government sector) rise as a result of the deficit without a corresponding increase in liabilities. It is this result that leads to the conclusion that that deficits increase net financial assets in the non-government sector.
What would happen if there were bond sales? All that happens is that the banks reserves are reduced by the bond sales but this does not reduce the deposits created by the net spending. So net worth of the non-government sector is not altered. What is changed is the composition of the asset portfolio held in the non-government sector.
The only difference between the Treasury “borrowing from the central bank” and issuing debt to the private sector is that the central bank has to use different operations to pursue its policy interest rate target. If it debt is not issued to match the deficit then it has to either pay interest on excess reserves (which most central banks are doing now anyway) or let the target rate fall to zero (the Japan solution).
There is no difference to the impact of the deficits on net worth in the non-government sector.
Mainstream economists would say that by draining the reserves, the central bank has reduced the ability of banks to lend which then, via the money multiplier, expands the money supply.
However, reflect back on the errors of logic outlined above in the way the mainstream macroeconomists use the GBC:
- Building bank reserves does not increase the ability of the banks to lend. They do not need reserves to lend. Banks will lend if there are credit-worthy customers irrespective of the operations that accompany government net spending. Please read the following blog – Building bank reserves will not expand credit.
- The money multiplier fails to adequately describe the credit creation process – that is, the way in which banks make loans.
- Inflation is caused by aggregate demand growing faster than real output capacity. The reserve position of the banks is not functionally related with that process. Please read the following blog – Building bank reserves is not inflationary – for further discussion.
None of this leads to the conclusion that budget deficits do not carry an inflation risk. All components of aggregate demand carry an inflation risk if they become excessive, which can only be defined in terms of the relation between spending and productive capacity. It might be that in your example a 6 per cent deficit with full employment might be too expansionary.
There is nothing a priori that would suggest that. There is nothing intrinsically interesting about a 6 per cent deficit or a 2 per cent deficit. It all depends on the spending contributions from the three sectors (government, external and private domestic).
If a 6 per cent deficit – which is a flow of spending – was adding too much to aggregate demand (spending) at full employment and wasn’t just offsetting the demand drains coming from the external sector (trade deficit) and the private domestic sector (saving overall) – then MMT provides clear guidelines for policy.
As long as the government is happy with the political private-public mix of activity then it would cut back its deficit to avoid pushing demand into the inflation barrier. If it wanted more public activity and less private, it could do that by increasing taxes and deprive the private sector of disposable income.
But the assertion that the private placement of debt reduces the inflation risk is totally fallacious. It does not.
I hope this clarifies some of the burning issues that are current but very obscured by erroneous public commentary.
Budget deficits do matter because they are the way the government can add to aggregate demand and influence real output and employment levels. If they add too much nominal demand then the government will cause inflation. It is really very simple.
But the actual budget deficit outcome that is associated with maximum output and employment is of no significance. The only legitimate goal that government should pursue is the real goals – employment and output – focusing on some given (fiscal rule-driven) budget outcome will likely thwart that legitimate activity.
That is enough for today!