Saturday Quiz – November 16, 2013 – answers and discussion

Here are the answers with discussion for yesterday’s quiz. The information provided should help you work out why you missed a question or three! If you haven’t already done the Quiz from yesterday then have a go at it before you read the answers. I hope this helps you develop an understanding of modern monetary theory (MMT) and its application to macroeconomic thinking. Comments as usual welcome, especially if I have made an error.

Question 1:

Only one of the following propositions is possible for a nation over any given period (with all balances expressed as a per cent of GDP):

• An external deficit accompanied by a public surplus of equal size and the private domestic sector saving overall.
• An external deficit accompanied by a public surplus of equal size and the private domestic sector dis-saving overall.
• An external deficit accompanied by a larger public surplus and the private domestic sector saving overall.

The best answer is the second option – “An external deficit accompanied by a public surplus of equal size and the private domestic sector spending dis-saving overall”.

This is a question about the sectoral balances – the government budget balance, the external balance and the private domestic balance – that have to always add to zero because they are derived as an accounting identity from the national accounts.

To refresh your memory the balances are derived as follows. The basic income-expenditure model in macroeconomics can be viewed in (at least) two ways: (a) from the perspective of the sources of spending; and (b) from the perspective of the uses of the income produced. Bringing these two perspectives (of the same thing) together generates the sectoral balances.

From the sources perspective we write:

GDP = C + I + G + (X – M)

which says that total national income (GDP) is the sum of total final consumption spending (C), total private investment (I), total government spending (G) and net exports (X – M).

From the uses perspective, national income (GDP) can be used for:

GDP = C + S + T

which says that GDP (income) ultimately comes back to households who consume (C), save (S) or pay taxes (T) with it once all the distributions are made.

Equating these two perspectives we get:

C + S + T = GDP = C + I + G + (X – M)

So after simplification (but obeying the equation) we get the sectoral balances view of the national accounts.

(I – S) + (G – T) + (X – M) = 0

That is the three balances have to sum to zero. The sectoral balances derived are:

• The private domestic balance (I – S) – positive if in deficit, negative if in surplus.
• The Budget Deficit (G – T) – negative if in surplus, positive if in deficit.
• The Current Account balance (X – M) – positive if in surplus, negative if in deficit.

These balances are usually expressed as a per cent of GDP but that doesn’t alter the accounting rules that they sum to zero, it just means the balance to GDP ratios sum to zero.

A simplification is to add (I – S) + (X – M) and call it the non-government sector. Then you get the basic result that the government balance equals exactly $-for-$ (absolutely or as a per cent of GDP) the non-government balance (the sum of the private domestic and external balances).

This is also a basic rule derived from the national accounts and has to apply at all times.

The following Table represents the three options in percent of GDP terms. To aid interpretation remember that (I-S) > 0 means that the private domestic sector is spending more than they are earning; that (G-T) < 0 means that the government is running a surplus because T > G; and (X-M) < 0 means the external position is in deficit because imports are greater than exports.

The first two possibilities we might call A and B:

A. An external deficit accompanied by a public surplus of equal size and the private domestic sector spending saving overall.
B. An external deficit accompanied by a public surplus of equal size and the private domestic sector spending dis-saving overall.

So Option A says the private domestic sector is saving overall (spending less than they are earning), whereas Option B say the private domestic sector is dis-saving ((spending more than they are earning and going into increasing indebtedness). These options are captured in the first column of the Table. So the arithmetic example depicts an external sector deficit of 2 per cent of GDP and an offsetting budget surplus of 2 per cent of GDP.

You can see that the private sector balance is positive (that is, the sector is spending more than they are earning – Investment is greater than Saving – and has to be equal to 4 per cent of GDP.

Given that the only proposition that can be true is:

B: An external deficit accompanied by a public surplus of equal size and the private domestic sector spending dis-saving overall.

Column 2 in the Table captures Option C:

C: An external deficit accompanied by a larger public surplus and the private domestic sector spending saving overall.

So the current account deficit is equal to 2 per cent of GDP while the surplus is now larger at 3 per cent of GDP. You can see that the private domestic deficit rises to 5 per cent of GDP to satisfy the accounting rule that the balances sum to zero.

So what is the economic rationale for this result?

If the nation is running an external deficit it means that the contribution to aggregate demand from the external sector is negative – that is net drain of spending – dragging output down.

The external deficit also means that foreigners are increasing financial claims denominated in the local currency. Given that exports represent a real costs and imports a real benefit, the motivation for a nation running a net exports surplus (the exporting nation in this case) must be to accumulate financial claims (assets) denominated in the currency of the nation running the external deficit.

A fiscal surplus also means the government is spending less than it is “earning” and that puts a drag on aggregate demand and constrains the ability of the economy to grow.

In these circumstances, for income to be stable, the private domestic sector has to spend more than they earn.

You can see this by going back to the aggregate demand relations above. For those who like simple algebra we can manipulate the aggregate demand model to see this more clearly.

Y = GDP = C + I + G + (X – M)

which says that the total national income (Y or GDP) is the sum of total final consumption spending (C), total private investment (I), total government spending (G) and net exports (X – M).

So if the G is spending less than it is “earning” and the external sector is adding less income (X) than it is absorbing spending (M), then the other spending components must be greater than total income

The following blogs may be of further interest to you:

• Barnaby, better to walk before we run
• Stock-flow consistent macro models
• Norway and sectoral balances
• The OECD is at it again!

Question 2:

If in attempting to estimate the cyclical component of a government budget outcome we underestimate the potential capacity of an economy, we will conclude that the government’s discretionary fiscal position is less expansionary than it actually is.

The answer is False.

The implicit estimates of potential GDP that are produced by central banks, treasuries and other bodies like the IMF and the OECD are typically too pessimistic.

The reason is that they typically use the NAIRU to compute the “full capacity” or potential level of output which is then used as a benchmark to compare actual output against. The reason? To determine whether there is a positive output gap (actual output below potential output) or a negative output gap (actual output above potential output).

These measurements are then used to decompose the actual budget outcome at any point in time into structural and cyclical budget balances. The budget components are adjusted to what they would be at the potential or full capacity level of output.

So if the economy is operating below capacity then tax revenue would be below its potential level and welfare spending would be above. In other words, the budget balance would be smaller at potential output relative to its current value if the economy was operating below full capacity. The adjustments would work in reverse should the economy be operating above full capacity.

If the budget is in deficit when computed at the “full employment” or potential output level, then we call this a structural deficit and it means that the overall impact of discretionary fiscal policy is expansionary irrespective of what the actual budget outcome is presently. If it is in surplus, then we have a structural surplus and it means that the overall impact of discretionary fiscal policy is contractionary irrespective of what the actual budget outcome is presently.

So you could have a downturn which drives the budget into a deficit but the underlying structural position could be contractionary (that is, a surplus). And vice versa.

The difference between the actual budget outcome and the structural component is then considered to be the cyclical budget outcome and it arises because the economy is deviating from its potential.

As you can see, the estimation of the benchmark is thus a crucial component in the decomposition of the budget outcome and the interpretation we place on the fiscal policy stance.

If the benchmark (potential output) is estimated to be below what it truly is, then a sluggish economy will be closer to potential than if you used the true full employment level of output. Under these circumstances, one would conclude that the fiscal stance was more expansionary than it truly was.

This is very important because the political pressures may then lead to discretionary cut backs to “reign in the structural deficit” even though it is highly possible that at that point in time, the structural component is actually in surplus and therefore constraining growth.

The mainstream methodology involved in estimating potential output almost always uses some notion of a NAIRU which itself is unobserved. The NAIRU estimates produced by various agencies (OECD, IMF etc) always inflate the true full employment unemployment rate and completely ignore underemployment, which has risen sharply over the last 20 years.

The following graph is for Australia but it broadly representative of the types of constructs we are dealing with. It plots three different measures of labour market tightness:

• The gap between the actual unemployment rate and the Australian Treasury estimate of the NAIRU (blue line), which is interpreted as estimating full employment when the gap is zero (cutting the horizontal axis).
• The gap between the actual unemployment rate and a 2 per cent full employment rate (red line), again would indicate full employment if the line cut the horizontal axis.
• The gap between the broad labour underutilisation rate published by the ABS (available HERE), which takes into account underemployment and our 2 per cent full employment rate (green line).

Some might ask why would we assume that 2 per cent unemployment rate is a true full employment level? We know that unemployment will always be non-zero because of frictions – people leaving jobs and reconnecting with other employers. This component is somewhere around 2 per cent. The other components of unemployment which economists define are seasonal, structural and demand-deficient. Seasonal unemployment is tied up with frictional and likely to be small.

The concept of structural unemployment is vexed. I actually don’t think it exists because ultimately comes down to demand-deficient. The concept is biased towards a view that only private market employment are real jobs and so if the market doesn’t want a particular skill group or does not choose to provide work in a particular geographic area then the mis-match unemployment is structural.

The problem is that often there are unemployed workers in areas where employers claim there are skills shortages. The firms will not employ these workers and offer them training opportunities within the paid work environment because they exercise discrimination. So what is actually considered structural is just a reflection of employer prejudice and an unwillingness to extend training opportunities to some cohorts of workers.

Also, the government can always generate enough demand to provide jobs to all in every area should it choose. So ultimately, any unemployment that looks like it is “structural” is in fact due to a lack of demand.

So there is no reason why any economy cannot get their unemployment rate down to 2 per cent.

Given that, the NAIRU estimates not only inflate the true full employment unemployment rate but also completely ignore the underemployment, which has risen sharply over the last 20 years.

In the June quarter 2006 the Australian NAIRU gap was zero whereas the actual unemployment rate was still 2.78 per cent above the full employment unemployment rate. The thick red vertical line depicts this distance.

However, if we considered the labour market slack in terms of the broad labour underutilisation rate published by the ABS then the gap would be considerably larger – a staggering 9.4 per cent. Thus you have to sum the red and green vertical lines shown at June 2008 for illustrative purposes.

This means that the Australian Treasury are providing advice to the Federal government claiming that in June 2008 the Australian economy was at full employment when it is highly likely that there was upwards of 9 per cent of willing labour resources being wasted. That is how bad the NAIRU period has been for policy advice.

But in relation to this question, in June 2008, the Australian Treasury would have classified all of the federal budget balance in that quarter as being structural given that the cycle was considered to be at the peak (what they term full employment).

However, if we define the true full employment level was at 2 per cent unemployment and zero underemployment, then you can see that, in fact, the Australian economy would have been operating well below the full employment level and so there would have been a significant cyclical component being reflected in the budget balance.

Given the federal budget in June 2008 was in surplus the Treasury would have classified this as mildly contractionary whereas in fact the Commonwealth government was running a highly contractionary fiscal position which was preventing the economy from generating a greater number of jobs.

The point is that by reducing the potential GDP estimates (by inflating the estimate of full employment unemployment) the structural deficits always contain some cyclical component and suggest that the discreationary policy choice is more expansionary than what it truly is when calibrated against a more meaningful potential GDP measure.

The following blogs may be of further interest to you:

• The dreaded NAIRU is still about!
• Structural deficits – the great con job!
• Structural deficits and automatic stabilisers
• Another economics department to close

Question 3:

Governments concerned with their public debt ratio should encourage growth because the debt ratio falls once economic growth resumes.

The answer is False.

The primary deficit may not fall when economic growth is positive if discretionary policy changes offset the declining net spending as tax revenue increases and welfare payments fall (the automatic stabilisation).

Under current institutional arrangements, governments around the world voluntarily issue debt into the private bond markets to match $-for-$ their net spending flows in each period. A sovereign government within a fiat currency system does not have to issue any debt and could run continuous budget deficits (that is, forever) with a zero public debt.

The reason they is covered in the following blogs – On voluntary constraints that undermine public purpose.

The framework for considering this question is provided by the accounting relationship linking the budget flows (spending, taxation and interest servicing) with relevant stocks (base money and government bonds).

This framework has been interpreted by the mainstream macroeconomists as constituting an a priori financial constraint on government spending (more on this soon) and by proponents of Modern Monetary Theory (MMT) as an ex post accounting relationship that has to be true in a stock-flow consistent macro model but which carries no particular import other than to measure the changes in stocks between periods. These changes are also not particularly significant within MMT given that a sovereign government is never revenue constrained because it is the monopoly issuer of the currency.

To understand the difference in viewpoint we might usefully start with the mainstream view. The way the mainstream macroeconomics textbooks build this narrative is to draw an analogy between the household and the sovereign government and to assert that the microeconomic constraints that are imposed on individual or household choices apply equally without qualification to the government. The framework for analysing these choices has been called the government budget constraint (GBC) in the literature.

The GBC is in fact an accounting statement relating government spending and taxation to stocks of debt and high powered money. However, the accounting character is downplayed and instead it is presented by mainstream economists as an a priori financial constraint that has to be obeyed. So immediately they shift, without explanation, from an ex post sum that has to be true because it is an accounting identity, to an alleged behavioural constraint on government action.

The GBC is always true ex post but never represents an a priori financial constraint for a sovereign government running a flexible-exchange rate non-convertible currency. That is, the parity between its currency and other currencies floats and the the government does not guarantee to convert the unit of account (the currency) into anything else of value (like gold or silver).

This literature emerged in the 1960s during a period when the neo-classical microeconomists were trying to gain control of the macroeconomic policy agenda by undermining the theoretical validity of the, then, dominant Keynesian macroeconomics. There was nothing particularly progressive about the macroeconomics of the day which is known as Keynesian although as I explain in this blog – Those bad Keynesians are to blame – that is a bit of a misnomer.

Just as an individual or a household is conceived in orthodox microeconomic theory to maximise utility (real income) subject to their budget constraints, this emerging approach also constructed the government as being constrained by a budget or “financing” constraint. Accordingly, they developed an analytical framework whereby the budget deficits had stock implications – this is the so-called GBC.

So within this model, taxes are conceived as providing the funds to the government to allow it to spend. Further, this approach asserts that any excess in government spending over taxation receipts then has to be “financed” in two ways: (a) by borrowing from the public; and (b) by printing money.

You can see that the approach is a gold standard approach where the quantity of “money” in circulation is proportional (via a fixed exchange price) to the stock of gold that a nation holds at any point in time. So if the government wants to spend more it has to take money off the non-government sector either via taxation of bond-issuance.

However, in a fiat currency system, the mainstream analogy between the household and the government is flawed at the most elemental level. The household must work out the financing before it can spend. The household cannot spend first. The government can spend first and ultimately does not have to worry about financing such expenditure.

From a policy perspective, they believed (via the flawed Quantity Theory of Money) that “printing money” would be inflationary (even though governments do not spend by printing money anyway. So they recommended that deficits be covered by debt-issuance, which they then claimed would increase interest rates by increasing demand for scarce savings and crowd out private investment. All sorts of variations on this nonsense has appeared ranging from the moderate Keynesians (and some Post Keynesians) who claim the “financial crowding out” (via interest rate increases) is moderate to the extreme conservatives who say it is 100 per cent (that is, no output increase accompanies government spending).

So the GBC is the mainstream macroeconomics framework for analysing these “financing” choices and it says that the budget deficit in year t is equal to the change in government debt (ΔB) over year t plus the change in high powered money (ΔH) over year t. If we think of this in real terms (rather than monetary terms), the mathematical expression of this is written as:

which you can read in English as saying that Budget deficit (BD) = Government spending (G) – Tax receipts (T) + Government interest payments (rB_t-1), all in real terms.

However, this is merely an accounting statement. It has to be true if things have been added and subtracted properly in accounting for the dealings between the government and non-government sectors.

In mainstream economics, money creation is erroneously depicted as the government asking the central bank to buy treasury bonds which the central bank in return then prints money. The government then spends this money. This is called debt monetisation and we have shown in the Deficits 101 series how this conception is incorrect. Anyway, the mainstream claims that if the government is willing to increase the money growth rate it can finance a growing deficit but also inflation because there will be too much money chasing too few goods! But an economy constrained by deficient demand (defined as demand below the full employment level) responds to a nominal impulse by expanding real output not prices.

But because they believe that inflation is inevitable if “printing money” occurs, mainstream economists recommend that governments use debt issuance to “finance” their deficits. But then they scream that this will merely require higher future taxes. Why should taxes have to be increased?

Well the textbooks are full of elaborate models of debt pay-back, debt stabilisation etc which all “prove” (not!) that the legacy of past deficits is higher debt and to stabilise the debt, the government must eliminate the deficit which means it must then run a primary surplus equal to interest payments on the existing debt.

Nothing is included about the swings and roundabouts provided by the automatic stabilisers as the results of the deficits stimulate private activity and welfare spending drops and tax revenue rises automatically in line with the increased economic growth. Most orthodox models are based on the assumption of full employment anyway, which makes them nonsensical depictions of the real world.

More sophisticated mainstream analyses focus on the ratio of debt to GDP rather than the level of debt per se. They come up with the following equation – nothing that they now disregard the obvious opportunity presented to the government via ΔH. So in the following model all net public spending is covered by new debt-issuance (even though in a fiat currency system no such financing is required).

Accordingly, the change in the public debt ratio is:

The change in the debt ratio is the sum of two terms on the right-hand side: (a) the difference between the real interest rate (r) and the GDP growth rate (g) times the initial debt ratio; and (b) the ratio of the primary deficit (G-T) to GDP.

A growing economy can absorb more debt and keep the debt ratio constant. For example, if the primary deficit is zero, debt increases at a rate r but the debt ratio increases at r – g.

So a change in the change in the debt ratio is the sum of two terms on the right-hand side: (a) the difference between the real interest rate (r) and the GDP growth rate (g) times the initial debt ratio; and (b) the ratio of the primary deficit (G-T) to GDP.

As we noted a growing economy can absorb more debt and keep the debt ratio constant. For example, if the primary deficit is zero, debt increases at a rate r but the debt ratio increases at r – g.

Consider the following table which simulates two different scenarios. Case A shows a real interest rate of zero and a steadily increasing annual GDP growth rate across 10 years. The initial public debt ratio is 100 per cent (so well over the level Reinhart and Rogoff claim is the point of no return and insolvency is pending). The budget deficit is also simulated to be 5 per cent of GDP then reduces as the GDP growth induce the automatic stabilisers. It then reaches a steady 2 per cent per annum which might be sufficient to support the saving intentions of the non-government sector while still promoting steady economic growth.

You can see that the even with a continuous deficit, the public debt ratio declines steadily and would continue to do so as the growth continued. The central bank could of-course cut the nominal interest rate to speed the contraction in the debt ratio although I would not undertake that policy change for that reason.

In Case B we assume that the government stops issuing debt with everything else the same. The public debt ratio drops very quickly under this scenario.

However, should the real interest rate exceed the economic growth rate, then unless the primary budget balance offsets the rising interest payments as percent of GDP, then the public debt ratio will rise.

The only concern I would have in this situation does not relate to the rising ratio. Focusing on the cause should be the policy concern. If the real economy is faltering because interest rates are too high or more likely because the primary budget deficit is too low then the rising public debt ratio is just telling us that the central bank should drop interest rates or the treasury should increase the discretionary component of the budget.

In general though, the public debt ratio is a relatively uninteresting macroeconomic figure and should be disregarded. If the government is intent on promoting growth, then the primary deficit ratio and the public debt ratio will take care of themselves.

You may be interested in reading these blogs which have further information on this topic:

• On voluntary constraints that undermine public purpose
• Deficits 101 series
• Questions and answers 1
• Will we really pay higher taxes?
• Will we really pay higher interest rates?