I am often sent E-mails asking me to explain succinctly (what my other explanations are not!) how public deficits finance saving. What does it mean? How does it work in a macroeconomic system? What is the difference between automatic stabilisers and discretionary budget dynamics? What would have happened if the government had not have increased the growth in spending? All these sorts of questions. So this short blog – to make up for yesterday’s ridiculously long blog – will cover those issues. It should clear up any outstanding issues about why deficits are important to underwriting growth.
You might also like to read a blog entry on the UMKC site written in 2000 (recently recycled to keep reminding us) by Warren Mosler and Eileen Debold, both significant figures in bond markets and investment banking. They have a nice graph of the US economy which makes their point very well.
The following graph is of the Australian economy from 1974 to 2009 (the last year is just my estimate) and shows the Federal headline budget balance as a percentage of GDP (blue line except when in surplus where it is a red line) and Household savings as a percentage of disposable income (green line). Unlike others I consider a budget surplus to be “in the red” (that is a poor result) because it will always be the precursor to bad things to come despite the rhetoric to the contrary.
The shaded areas broadly correspond (approximate given this is annual data) with recessions (or significant downturns). If you study it carefully you can see that when the economy has gone into downturn, the budget deficit has increased. You will also see that just prior to the downturn, the budget has been in surplus or heading towards surplus. These dynamics are not accidental.
You can also see that over the period shown there has been a bias towards the budget deficit being smaller (trending down) as neo-liberal discretionary policy changes sought to diminish the size of government relative to the private sector. Over the same period, the household saving ratio (saving as a percentage of disposable income) has been steadily falling too. However, whenever there has been a sharp rise in the deficit (during the downturns shown) the household saving ratio has risen somewhat.
Now if we consider that in the post 1991 growth period, the financial engineers have been extremely active in pushing a diversity of financial debt products onto business firms and households. In the 1980s it was largely the private firms that became heavily leveraged but by the 1990s and beyond it was the household sector. Partly this has been due to the rising house prices that required increasing levels of debt to purchase. But it has also been strongly driven by diversifying the types of debt that households have been carrying.
So in this period we had margin debt emerging for the first time as household tried to live the shareholder dream whereby everyone becomes a capitalist!
By the 1996, the credit binge was accelerating and this allowed the conservative government to indulge in its obsession with budget surpluses. So even though the fiscal drag was destroying private sector purchasing power (and undermining their wealth as public debt was reduced), the increasing debt being borne by the household sector propped up spending sufficiently to allow tax revenue to rise faster than outlays. The result was a sequence of surpluses.
Ultimately, the household sector reached the point (around 2001-2002) where the saving ratio became negative – that is, the household sector was dis-saving overall. The increased precariousness of household finances was evident during this time. Even though GDP growth continued the central bank was very reluctant to increase interest rates by more than 25 basis points at a time because they became scared that there would be mass insolvency. Statistics like forced mortgagee sales rose quickly during this period.
So the only way the previous government was able to run so many surpluses in a row (the red segments) was because the non-government sector (principally the household sector) was living beyond its means – increasingly and funding that excess with rising levels of debt. If the financial engineers had not been so successful for so long, then you would have had a 1991 type recession much earlier following the surpluses.
In that period, the surpluses were not supported by a debt-binge (or rather the debt-binge was short-lived and choked by very high interest rates) and so the Keating surpluses of the late 1980s quickly became huge deficits as the automatic stabilisers put a floor under the economy. The costs of that surplus period are there for all to see – the prolonged and very serious 1991-1993 recession.
In the recent period, it was clear that the household sector could not dis-save indefinitely and eventually households would start to repair their fragile financial positions by increasing saving. Once they did that, the economy was always going to crash under the fiscal drag coming from the surpluses.
You can see now that as the household saving ratio has risen sharply so has the budget deficit. It more or less has to respond in that way because of the automatic stabilisers. However, in Australia’s case the federal government has also accelerated the move into deficit with two rather substantial and early fiscal stimulus packages.
What these packages achieve is simple. They provide a spending stimulus as the private sector reduces its spending and seeks to save. The net spending from the government allows production levels (and sales) to be higher than they otherwise would be in the face of a private sector spending contraction. The higher than otherwise output and employment levels allow the income levels in the economy to remain higher than they would otherwise be and this supports the private saving intentions (savings is a function of national income!).
It is in this sense that modern monetary theorists say that net public spending “finances” non-government saving. It provides the spending boost which allow income to remain higher than otherwise which in turn provides the capacity for the private sector to save.
Had the deficits not been rising as quickly, then the output loss would have been greater and the rise in private saving actually realised would have been smaller (because GDP would be lower). But this situation would continue to drive the deficits up via the automatic stabilisers but the government would have very little to show for the higher net spending. It would have worse unemployment and a longer recession to deal with.
The point is that it is typical for the household sector to save a proportion of disposable income. The dis-saving period is very atypical in our history. In that case, with a current account deficit, the only way that the domestic private sector can realise its desires to save is if the government is in deficit. We should consider deficits to be normal, typical and productive aspects of economic life.
They provide the spending capacity to underpin high levels of employment and high levels of private saving – both goals that allow for sustainable growth and reduced poverty.
As an aside, the Australian graph is not as obvious as the US graph because we have more noise from our external sector. We are much more open than the US economy which is relatively closed. To give you an idea of the external sector the second graph is the Current Account deficit as a percentage of GDP. You can see it is always in deficit although there is no obvious trend – which means that we can consider it a relatively constant source of foreign savings being made available to our economy over this period.
The way to think of this is that the excess of imports over exports (taking into account invisibles) is possible because foreigners want to accumulate financial assets denominated in AUD and are willing to sacrifice more real wealth (goods and services) that they receive in return to accomplish this investment strategy. Our real terms of trade have therefore been firmly in our favour over this period. Most mainstream economists will tell you this is a demonstration of us “living beyond our means” and that the foreigners are financing this excess. When you think of it correctly – in a modern monetary sense – it is us who are financing their investment strategies and in return we are extracting a surplus of real goods and services from them. Sounds to me as we win out in real terms.
The second set of graphs are from a simulated (very simple) macroeconomic model I have to demonstrate things. It assumes inflation and the exchange rate are constant which in this sort of environment is a reasonable assumption. It assumes that consumers save 20 per cent of every dollar they earn and government taxes a flat 20 per cent in the dollar. For every dollar of production generated, 20 cents go out to imports. Export demand is also assumed constant to isolate that from the analysis. So imports rise with income (GDP)
Potential GDP is assumed to grow at 2 per cent per year. The gap between GDP and Potential GDP will give some indication of the change in unemployment although the labour market is implicit only. The wider the gap the worse the economy is performing and the higher will be unemployment. The top graph in each pair shows this relationship.
The bottom graph in each pair shows the sectoral balances – the Budget balance as a % of GDP (surplus +, deficit -); the Private sector balance (the gap between saving and investment) as a % of GDP; and the Current account balance (exports minus imports) as a % of GDP. The three balances as a matter of national accounting have to add up to zero. If we summed the external and private balances we would get the non-government balance (broadly) and that would equal the mirror image of the government balance (the budget balance).
Simulation One – Government spending grows by 1 per cent per period
In the first simulation, the economy is nearly at full capacity but private spending is in excess of income and the budget balance is in surplus. In this environment the private sector is increasing its indebtedness to maintain spending growth. This is an unsustainable position and eventually the private sector will reduce its deficit position. We model that here by private investment (a component of aggregate demand) contracting by 10 units for each of the first 6 periods and then bottoming out for the last four periods of the 10-period cycle shown.
The government reacts immediately and increases real spending by 1 per cent per period (from an initial 2 per cent growth rate). You can see that while GDP falls initially eventually the net spending stimulus turns the economy around and it starts heading back into growth. I could have clearly tweaked the results to get the growth happening earlier and stronger but the dynamics would have been the same. The simple fact is that the deficit in this economy is still not large enough.
The deficits though help the private balance move into surplus (as desired by their discretionary spending choices).
If the government had not have increased the discretionary component of the budget (increasing spending each period) the drop in investment would have had much more dire circumstances and the income adjustments would have still sent the budget into deficit (a bit more slowly) and provided the funds to allow the private balance to go into surplus. This is shown in the second simulation.
Simulation Two – Government spending growth constant per period
In this case, the real government spending growth remains a constant 2 per cent per period and the tax rate remains at 20 per cent (so no discretionary fiscal policy change is modelled). You can see that the deterioration in the economy as the private sector withdraws spending is severe and the weak growth that emerges after period 7 comes exclusively from the automatic stabilisers (modelled here as the declining tax revenue as GDP falls) and the declining imports (as GDP falls). It is a parlous outcome.
So the automatic stabilisers provide a floor that the economy will not fall below but the capacity underutilisation in this second case is much higher and so the implied unemployment is much higher. The recession in this case would be much more protracted than in the first case.
The point of the simulations is not to capture everything about the macroeconomy. But what they will reliably represent are the dynamics between the major macroeconomic aggregates during a typical recession and early recovery period. I could have made them much more complicated to bring in other detail but the message would be the same.
The fiscal balance is essential to stabilise output when the non-government sector is in retreat. The economy will also be in better shape if the budget balance actively provides the stimulus necessary to support output when the non-government sector is desiring to increase its saving. The more aggressive is fiscal policy as non-government spending is in decline the more quickly GDP will recover and the higher will be the savings volume in the private sector.
If fiscal policy is reactive only (that is, reliant on automatic stabilisers) then the private saving balance will not increase as much because the income adjustments that make it possible for the private sector to save yet still allow the economy to grow will not be funded as strongly by the fiscal sector.
Digression: ABS to launch new employment series
The Australian Bureau of Statistics will tomorrow publish a new labour market measure – aggregate monthly hours worked – when it releases the monthly Labour Force data. Apparently, the data will be backdated to July 1985. The measure will provide a richer measure of labour demand than just employment by persons. Among other things, it will allow analysts such as me to work out how much labour hoarding is going on over the business cycle.
Digression: Two interesting phone calls this week
On Monday I received a call from Peter Andrews who is the Natural Sequence Farming person that you may have seen recently (and in the past) on Australian Story 2005 and Australian Story 2009 Part 1 and Australian Story 2009 Part 2.
He has created permaculture havens in deeply drought-affected areas of our continent using skill and understanding. A lateral thinker no doubt.
He wanted to discuss forming a network of like minds to combine my interest in job creation and land rehabilitation with his approaches to permaculture. Further discussions are pending.
Then today Opposition leader Malcolm Turnbull rang me to discuss these matters. Seems he is keen to develop policy in this area. He had met with Peter Andrews last week. Anyway, we had an interesting chat and exchanged (non-fake) E-mails. More to report later maybe. It would be ironic if the Federal Opposition became the champions for large-scale public sector job creation while the Labor Government sat on its tails and were content with work experience and wasteful training programs while unemployment soared.
For overseas readers, the Opposition leader is currently embroiled in a fake E-mail scandal where a senior Treasury official invented an E-mail purporting to reveal impropriety by the Prime Minister and induced the Opposition to use the fake E-mail as a basis for calling for the Government to resign. It has turned pear-shaped for the Opposition beyond imagination.