Budget surpluses are not national saving – redux

I was reading several older papers from the 1990s today as part of a project I am working on where I track predictions that leading mainstream economists were making at the time about the evolution of national and global economies. It is a very interesting exercise to build the narratives that were popular at an earlier time and then consider how far the economists got things right. I have noted that there has been some debate out in blog-land about who predicted the failure of the Euro. I am less interested in documenting which person was the first or the second – there were many who saw the design flaws from the inception and could extrapolate what they would mean if a negative shock occurred. Modern Monetary Theory (MMT) economists were among them. I am more interested in groupthink (at the paradigm level) and how the failed predictions can be used to demonstrate the inapplicability of a certain body of theory. That is, what can we learn from the failure of mainstream economists in general to see the crisis coming (and being in denial now of what the solution is). In this blog I consider a part of the thinking that explains why my profession proved to be unreliable in this regard. I renamed this blog – appending it with the term redux because on March 23rd, 2009 – I wrote a blog – Budget surpluses are not national saving.

I re-read a paper today – What Do Budget Deficits Do? – by Laurence Ball (Johns Hopkins University) and Greg Mankiw (Harvard University). It was presented at the 1995 Economic Symposium – Budget Deficits and Debt: Issues and Options – staged by the Federal Reserve Bank of Kansas City (an annual event). I should add at the outset that I have had some dealings with Laurence Ball in the past (work-related) and he is a thoroughly nice person!

The paper seeks to analyse the impact of budget deficits on the economy and the authors say that their:

The analysis follows the conventional wisdom as captured, for example, in most undergraduate textbooks. In our view, the conventional wisdom in this area is mostly on the right track.

In other words, they present the standard fare that macroeconomics students receive each day in universities around the world. The curriculum is more about mis-education than the positive. Through various sleights of hand it trains the students’ minds to ignore the essential and build a logic system that is internally consistent but, ultimately erroneous in terms of its applicability to the actual monetary system that we live in.

The paper is quite instructive because laid out before us is a good example, which can be used to demonstrate how Modern Monetary Theory (MMT) differs from the mainstream and can legitimately lay claim to providing a more valid representation of the way the economy actually operates.

We are assuming the general case of a government that issues its own currency (through some arrangement involving the central bank and the treasury) and floats that currency on international markets.

The authors begin by analysis the “immediate effects of budget deficits” and say that these impacts:

… all follow from a single initial effect: deficits reduce national saving. National saving is the sum of private saving (the after-tax income that households save rather than consume) and public saving (the tax revenue that the government saves rather than spends). When the government runs a budget deficit, public saving is negative, which reduces national saving below private saving.

Which is the first false premise and sleight of hand.

MMT attempts to address the confusion that is abroad about the concept of national saving. An intuitive understanding of the term – saving – is built on the notion that a household saves when it spends less than you earns after tax (that is, the proportion of disposable income that is not spent on consumption).

That understanding is correct. But then intuition fails if this is applied to the concept of a budget surplus. That application would conclude – as Ball and Mankiw opine that budget surpluses add to national saving because they arise when the government spends less than it earns.

The problem is that a sovereign government does not save when it runs a budget surplus.

What sense does it make to say that the government is saving in the currency that it issues? Households save to increase their capacity to spend in the future. Accordingly, they sacrifice consumption today to enhance their consumption possibilities in the future (by earning interest on the saving and hoping that inflation does not erode the purchasing power of their financial assets).

How can this logic apply to the issuer of the currency who can spend at any time it chooses? Does a budget surplus increase the capacity of the government to spend in the future? Clearly not.

As a matter of national accounting, a government budget deficit adds net financial assets (adding to non-government savings) available to the private sector and a budget surplus has the opposite effect. The last point requires further explanation as it is crucial to understanding MMT.

While typically obfuscated in standard textbook treatments, at the heart of national income accounting is an identity – the government deficit (surplus) equals the non-government surplus (deficit). The only entity that can provide the non-government sector with net financial assets (net savings denominated in the currency of issue) and thereby simultaneously accommodate any net desire to save (financial assets) and thus eliminate unemployment is the currency monopolist – the government.

It does this by net spending – that is, running deficits. Additionally, and contrary to mainstream rhetoric, yet ironically, necessarily consistent with national income accounting, the systematic pursuit of government budget surpluses is dollar-for-dollar manifested as declines in non-government savings.

A simple example helps reinforce these points. Suppose the economy is populated by two people, one being government and the other deemed to be the private (non-government) sector. We abstract from the distinction between the external and private domestic sectors here – which mostly only involved distributional considerations anyway.

If the government runs a balanced budget (spends 100 dollars and taxes 100 dollars) then private accumulation of fiat currency (savings) is zero in that period and the private budget is also balanced.

Say the government spends 120 and taxes remain at 100, then private saving is 20 dollars which can accumulate as financial assets. In the first instance, they would be sitting as a 20 dollar bank deposit have been created by the government to cover its additional expenses. The government deficit of 20 is exactly the private savings of 20.

If the government continued in this vein, accumulated private savings would equal the cumulative budget deficits. The government may decide to issue an interest-bearing bond to encourage saving but operationally it does not have to do this to finance its deficit. If the savers transfer their deposits into bonds their overall saving is not altered and it has no implications for the government’s capacity to spend. It has the advantage for savers that they now also enjoy an income flow from their saving.

However, should government decide to run a surplus (say spend 80 and tax 100) then the private sector would owe the government a net tax payment of 20 dollars and would need to sell something back to the government to get the needed funds. The result is the government generally buys back some bonds it had previously sold.

The net funding needs of the non-government sector automatically elicit this correct response from government via interest rate signals. Either way accumulated private saving is reduced dollar-for-dollar when there is a government surplus.

So it is clear that the government surplus has two negative effects for the private sector: (a) the stock of financial assets (money or bonds) held by the private sector, which represents its wealth, falls; and (b) private disposable income also falls in line with the net taxation impost.

Some may retort that government bond purchases provide the private wealth-holder with cash. That is true but the liquidation of wealth is driven by the shortage of cash in the private sector arising from tax demands exceeding income. The cash from the bond sales pays the Government’s net tax bill. The result is exactly the same when expanding this example by allowing for private income generation and a banking sector.

From the example above, and further recognising that currency plus reserves (the monetary base) plus outstanding government securities constitutes net financial assets of the non-government sector, the fact that the non-government sector is dependent on the government to provide funds for both its desired net savings and payment of taxes to the government becomes a matter of accounting.

This understanding reinforces why the pursuit of government budget surpluses will be contractionary. Pursuing budget surpluses is necessarily equivalent to the pursuit of non-government sector deficits. They are two sides of the same coin.

The decreasing levels of net savings financing the government surplus increasingly leverage the private sector and the deteriorating debt to income ratios will eventually see the system succumb to ongoing demand-draining fiscal drag through a slow-down in real activity.

Once we understand the accounting relationships it is easy to reject the argument that budget surpluses represents “public saving”, which can be used to fund future public expenditure. Public surpluses do not create a cache of money that can be spent later. National governments spend by crediting a reserve account in the banking system. The credits do not “come from anywhere”, as, for example, gold coins would have had to come from somewhere. It is accounted for but that is a different issue. Likewise, payments to government reduce reserve balances. Those payments do not “go anywhere” but are merely accounted for.

There is an elaborate institutional structure in place to obsfucate the true nature of these transactions. But in an accounting sense, when there is a budget surplus, then base money and/or private wealth is destroyed. The opposite is the case for budget deficits.

Please read my blogs – Deficit spending 101 – Part 1 – Deficit spending 101 – Part 2 – Deficit spending 101 – Part 3Please read my blog – for more background discussion on this point.

Once we understand that logic then the presentation by Ball and Mankiw can be see in a different light.

In asking the question “(h)ow does lower national saving affect the economy”, they use a national income model which they say comprises:

… some simple (and irrefutable) accounting identities.

I agree that national accounting identities (things that have add up by definition) are irrefutable. But they tell us very little unless we have an understanding of the underlying dynamic forces that produce the numbers that are accounted for. In other words, we need to apply theories to the accounting structures to make any sense of them (beyond meagre accounting) and to conduct inference from their use.

Causality is beyond accounting.

Here is their logic – and I invite you to spot the sleight of hand.

They say:

Letting Y denote gross domestic product, T taxes, C consumption, and G government purchases, then private saving is Y-T-C, and public saving is T-G. Adding these yields national saving, S:

S = Y – C – G.

National saving is current income not used immediately to finance consumption by households or purchases by the government.

The second crucial accounting identity is the one that divides GDP into four types of spending:

Y = C + I + G + NX.

Output Y is the sum of consumption C, investment I, government purchases G, and net exports NX. Substituting this expression for Y into the previous equation for national saving yields

S = I + NX.

From which they conclude that the last equation (National Savings equals Private Investment plus Net Exports) means that “when budget deficits reduce national saving, they must reduce investment, reduce net exports, or both”.

The sleight of hand is in the aggregation of the budget balance (T-G) and private saving (S = Y – T – C). Both equations are fine in an accounting sense.

Household saving is what is left over after the government has levied taxes (T) on total income (Y) and households have then consumed (C). Further the budget balance is equal $-for-$ to what it gains in revenue (T) minus what it spends (G), where presumably G includes transfers and interest payments in addition to public consumption and public capital formation (investment).

So the intermediate step from that which has more meaning in a modern monetary economy would be (with S now being the more specific term – household or private saving):

(1)      S = Y – T – C

and the government budget surplus is:

(2)      BS = T – G

Adding (1) and (2) to get the sum of private saving and the budget outcome (which is hidden in Ball and Mankiw’s term – S) we get:

(3)      S + BS = Y – T – C + T – G

Simplifying (noting that T cancels out) we get:

(4)      S + BS = Y – C – G

which is the first Ball and Mankiw equation expanded to recognise that private saving is very different in impact from a budget surplus (as explained above).

Their “second crucial accounting identity” is the standard Keynesian expenditure-income relationships which denotes the sources of spending that combine to generate GDP (national output or income):

(5)      Y = C + I + G + NX

That is uncontroversial.

So now we substitute (4) into (5) to get a different version of the relationship between saving and the spending aggregates. The steps are:

(6a)      Y = C + I + G + NX

(6b)      Y – C – G = I + NX

From (4):

(6c)      S + BS = I + NX

which we re-arrange to get:

(6)      S = I + NX – BS

Compare that to Ball and Mankiw’s last equation S = I + NX, which conflates the budget balance and household saving and fudges that sum as national saving.

Equation (6) is a very different configuration in terms of the narrative that you can now tell.

Increasing investment and net exports adds to aggregate demand and increases national income (GDP). Given that household saving is a positive function of national income (the higher is national income other things equal the larger will be consumption and saving) variables that augment aggregate pending impact positively on household or private saving.

Conversely, a budget surplus (T-G > 0) drains aggregate demand – by the government taking more purchasing power out of the economy than it adds back via spending – and other things equal – would result in a reduction in national income (GDP) and a decline in saving.

Further, if private households decided to save more out of disposable income (that is, the marginal propensity to consume fell) then from Equation (6a) this would manifest as an initial decline in consumption which would see national income decline. That may push the budget surplus towards or into deficit (via the lost tax revenue and/or increased welfare spending).

It is also likely to boost net exports (as long as there wasn’t a global recession) because import spending is a positive function of national income.

If the spending contraction was significant enough, business firms might also decide to reduce investment spending while they re-assess the future revenue prospects arising from the sale of goods and services produced by capital (plant and equipment etc) that was created by the flow of investment spending. Certainly accelerator models of investment would predict that as would most theories of sentiment.

The same impacts on investment and net exports could arise (in a causal sense) from a government trying to boost the budget surplus through the imposition of fiscal austerity. That is, in Ball and Mankiw’s erroneous terminology, a rise in national saving would damage investment and net exports.

The trick is in understanding that the causality in Equations (6) and (5), for example, is bi-directional and both sides of the accounting identies feedback on each other via income changes driven by the expenditure multiplier.

Ball and Mankiw try to explain their logic in some more detail. Focusing on their last equation (S = I + NX) they say:

At first, some of these conclusions may appear mysterious. Business firms choose the economy’s level of investment, and domestic and foreign consumers choose net exports. These decisions may seem independent of the political decisions that determine the budget deficit. If the government decides to run a deficit, what forces induce firms to invest less and foreigners to buy fewer domestic products?

They then introduce the second major false premise by claiming that “these changes are brought about by interest rates and exchange rates”.

First, they state that:

Interest rates are determined in the market for loans, where savers lend money to households and firms who desire funds to invest. A decline in national saving reduces the supply of loans available to private borrowers, which pushes up the interest rate (the price of a loan). Faced with higher interest rates, households and firms choose to reduce investment.

Which is the loanable funds theory from the C19th, which was thoroughly discredited by Keynes and his accomplices in the 1930s. In modern parlance Ball and Mankiw are rehearsing the mainstream notion of financial crowding out.

Underpinning of the crowding out hypothesis is the old Classical theory of loanable funds, which is an aggregate construction of the way financial markets work in mainstream macroeconomic thinking.

Mainstream textbook writers (for example, Mankiw) assume that it is reasonable to represent the financial system to his students as the “market for loanable funds” where “all savers go to this market to deposit their savings, and all borrowers go to this market to get their loans. In this market, there is one interest rate, which is both the return to saving and the cost of borrowing.”

The original conception was designed to explain how aggregate demand could never fall short of aggregate supply because interest rate adjustments would always bring investment and saving into equality. If consumption fell, then saving would rise and this would not lead to an oversupply of goods because investment (capital goods production) would rise in proportion with saving.

So while the composition of output might change (workers would be shifted between the consumption goods sector to the capital goods sector), a full employment equilibrium was always maintained as long as price flexibility was not impeded. The interest rate became the vehicle to mediate saving and investment to ensure that there was never any gluts.

The supply of funds comes from those people who have some extra income they want to save and lend out. The demand for funds comes from households and firms who wish to borrow to invest (houses, factories, equipment etc). The interest rate is the price of the loan and the return on savings and thus the supply and demand curves (lines) take the shape they do.

This framework is then used to analyse fiscal policy impacts and the alleged negative consequences of budget deficits – the so-called financial crowding out – is derived.

The erroneous mainstream logic claims that investment falls when the government borrows to match its budget deficit – the borrowing allegedly increases competition for scarce private savings pushes up interest rates. The higher cost of funds crowds thus crowds out private borrowers who are trying to finance investment. This leads to the conclusion that given investment is important for long-run economic growth, government budget deficits reduce the economy’s growth rate.

It is clear that sovereign governments borrow from the private sector – for both stupid ideological reasons (pretending they are financially constrained so as to limit government spending in a political sense) and to facilitate central bank operations.

Two questions arise: Doesn’t this borrowing increase the claim on saving and reduce the “loanable funds” available for investors? Doesn’t the competition for saving push up the interest rates?

The answer to both questions is no! MMT does not claim that central bank interest rate hikes are not possible. There is also the possibility that rising interest rates reduce aggregate demand via the balance between expectations of future returns on investments and the cost of implementing the projects being changed by the rising interest rates.

But the Classical claims about financial crowding out (draining finite saving) are not based on these mechanisms. In fact, they assume that savings are finite and the government spending is financially constrained which means it has to seek “funding” in order to progress their fiscal plans. The result competition for the “finite” saving pool drives interest rates up and damages private spending.

A related theory, which is taught under the banner of IS-LM theory (in macroeconomic textbooks), assumes that the central bank can exogenously set the money supply. Then the rising income from the deficit spending pushes up money demand and this squeezes (real) interest rates up to clear the money market. This is the Bastard Keynesian approach to financial crowding out.

Neither theory is remotely correct and is not related to the fact that central banks push up interest rates up because they believe they should be fighting inflation and interest rate rises stifle aggregate demand.

From a macroeconomic flow of funds perspective, the funds (net financial assets in the form of reserves) that are the source of the capacity to purchase the public debt in the first place come from net government spending. Its what astute financial market players call “a wash”. The funds used to buy the government bonds come from the government!

Further, there is no finite pool of saving that is competed for. First, deficit spending generates income growth which generates higher saving. It is this way that MMT shows that deficit spending supports or “finances” private saving not the other way around. Refer back to the earlier discussion.

Second, the loanable funds doctrine is in denial of the way modern banking operates. When a bank extends a loan it simultaneously creates a deposit and any credit-worthy customer can typically get funds (although the hurdles that such a person has to jump through can range from not many to several depending on the risk averseness of the banks).

Reserves to support these loans are added later – that is, loans are never constrained in an aggregate sense by a “lack of reserves”.

Acknowledging the point that increased aggregate demand, in general, generates income and saving, Michel Kalecki showed us clearly that – “investment brings forth its own savings” – an insight that demolished classical loanable funds theory upon which the neo-liberal crowding out theory was originally conceived.

Ball and Mankiw then introduce a version of what is known as exchange rate crowding out:

Higher interest rates also affect the flow of capital across national boundaries. When domestic assets pay higher returns, they are more attractive to investors both at home and abroad. The increased demand for domestic assets affects the market for foreign currency: if a foreigner wants to buy a domestic bond, he must first acquire the domestic currency. Thus, a rise in interest rates increases the demand for the domestic currency in the market for foreign exchange, causing the currency to appreciate.

Although earlier they said that “To the extent that budget deficits increase the trade deficit (that is, reduce net exports), another effect follows immediately: budget deficits create a flow of assets abroad.”

It is clear that a depreciating exchange rate is largely stimulatory because it means the net spending injection from abroad is positive (or less negative depending on the state of the current account). An appreciating exchange rate is contractionary because export prices rise and import prices fall in local currency terms (usually – depending on the currencies that contracts are signed in).

But their argument hinges on the interest rate impact. If the central bank does not lift interest rates then the causality they suggest fails.

Overall, the evidence available does not support the hypothesis that budget deficits cause exchange rate appreciations. If anything, budget deficits are more associated with the opposite impact although economists have no convincing models which predict exchange rate movements.

In Australia’s case we have had a parity at 49 cents against the US dollar in early 2000s when the budget surplus was at record levels and more recently with a large absolute budget deficit the exchange rate has gone above parity against the US dollar.

The second part of the paper considers long-run effects of budget deficits – in terms of output and wealth.

They argue that budget deficits undermine the long-run growth path of the economy, which is:

… is determined by its productive capacity, which, in turn, is partly determined by its stock of capital. When deficits reduce investment, the capital stock grows more slowly than it otherwise would.

There is no convincing evidence that budget deficits are related to lower investment growth. There are problems with causality (so-called endogeneity problems) that haunt econometric models that attempt to demonstrate the point.

It is easy to show that at times a falling investment ratio is co-incident with a rising budget deficit. Does that prove that rising budget deficits cause the falling investment ratios?

Clearly not. Typically budget deficits will rise when aggregate spending (of which investment is a key fluctuating component) falls because of the automatic stabiliser effect – lower national income leading to lower tax revenue and rising welfare spending.

Similarly, when investment spending is strong, tax revenue booms and the budget balance can be reduced if not pushed into surplus on the back of the bouyant economy.

Thus, the causality typically is reverse to what the mainstream argue.

I have run out of time today to consider their additional arguments which are:

1. Budget deficits reduce national saving and increase the foreign ownership of domestic bonds, real estate, or equity which reduces the national income available to residents.

2. Budget deficits undermine marginal productivity of labour by reducing the capital stock and so real wages fall.

3. Budget deficits increase future tax rates and households reduce their spending in anticipation. They negate their own argument here by acknowledging that for most nations the “average growth rate over long periods has exceeded the average interest rate on government debt” (which means that the public debt ratio is stable and/or falling as long as their is growth) and the “debt will eventually become negligible relative to the size of the economy, even with no tax increases”. They also acknowledge that

Arguments 1 and 2 require the earlier logic to be sound. Clearly it is not. Please read my blog – Budget deficit basics – for more discussion on these points.

Conclusion

The paper is a classic example of the way the mainstream constructs their case against budget deficits. For those who want to arm themselves with the counter-argument you would do well to learn their arguments in detail and then apply MMT to refute them with an eye to the empirical world to provide the real world buttress or reality check.

I think you will find that the MMT version of events and relationships is more supportable in terms of the evidence than the Ball and Mankiw version.

Alternative Olympic Games Medal Tally

My Alternative Olympic Games Medal Tally is now active.

I update it early in the day and again around lunchtime when all the sports are concluded for the day.

That is enough for today!

(c) Copyright 2012 Bill Mitchell. All Rights Reserved.