Jan Hatzius On Wynne Godley On Odd Lots
Jan Hatzius of Goldman Sachs talks about how he drew inspiration from Wynne Godley in the latest Odd Lots podcast hosted by Joe Weisenthal and Tracy Alloway of Bloomberg.
He talks of how the private sector financial balance (or net lending) is an important thing to look at to predict crisis.
The UK Office Of National Statistics (ONS) is planning to publish new flow of funds statistics in the future. Till now, the ONS had just the SNA tables. Although the SNA has lots of details on stocks and flows, flow of funds gives more details.
ONS summary:
See our first combined set of experimental flow of funds statistics using new data sources providing improved breakdowns of how money moves around the UK economy
The article also has a discussion on sectoral balances which has the chart on top of this post.
The 2008 SNA explains this:
This is a continuation of a recent post at this blog, Public Debt And Current Account Deficits, in which I argued that the current account balance of payments affects the public debt.
A usual objection to the connection is that the two deficits—current account deficit and the budget deficit—although connected by an identity, don’t move together and in fact move in the opposite direction frequently. This point was raised by the blog Econbrower, yesterday.
The identity in question is:
NL = DEF + CAB
where, NL is the private sector net lending, DEF is the government’s deficit and CAB is the current account balance of payments (and is to a zeroth order approximation, exports less imports).
This is not a behavioural hypothesis but still a useful tool to build a narrative. Also, the causality connecting the identities is domestic demand and output at home and abroad.
Imagine, initially that NL is a small positive relative to GDP (for example, NL/GDP = 2%), Also remember that,
NL = Private Income − Private Expenditure
Now assume that private expenditure rises relative to private income. This will lead to higher GDP, a higher national income and a rise in imports because of income effects and hence a lower CAB. It will also lead to higher taxes because of higher income and hence will reduce the budget deficit, DEF, ceteris paribus.
So if the current account balance is in deficit, it would mean that the budget deficit and the current account deficit move in opposite directions.
That’s the theoretical basis for the empirical relationship. But that in itself isn’t the whole story. This is because the other balance—net lending, NL—has a life of its own. As is the case in the United States and several western countries, it turned negative once or twice in the 1990s the 2000s, and when the private sector’s debt rose, it made a sharp U-turn into the positive territory. The blue line in this graph:
Click the graph to see it on FRED.
So, if net lending reverts to its mean of staying positive, one can then conclude that the cumulative budget deficit, or the public debt is affected by cumulative current account deficits.
At any rate, the public debt shouldn’t be the main object of study. What’s more important is the international investment position. And it’s an identity that:
NIIP = cumulative CAB + Revaluations
where, NIIP, is the net international investment position.
A nation which runs current account deficits can become indebted to the rest of the world. IIP is the position of assets and liabilities of resident sectors of a nation. So, the net debt (the negative of NIIP) is the nation’s debt.
The above linked Econbrowser post brings in the complication of revaluations to deny the relationship between CAB and NIIP. But revaluations can’t save you for long.
In short, both public debt and NIIP depend on current account deficits.
Finally a weak analogy: if you play in the rain, you might enjoy it as well. But then if you get sick, you can’t say, “I felt so good playing in rains, so playing in the rain didn’t make me sick”. Saying the two deficits (current account and budget) move in opposite directions is an argument like that.
Yesterday, there was an article at Vox which takes issue with a statement from Donald Trump which connects the US public debt with current account deficits.
Trump ran his campaign on dividing people, is out to destroy health care and wants a regressive system of taxation and so should be resisted. At the same time, a blanket opposition is counterproductive, especially when it is an important matter.
Vox quotes Trump:
For many, many years the United States has suffered through massive trade deficits; that’s why we have $20 trillion in debt.
In response, Vox claims:
The US trade deficit refers to the fact that the US imports more from the world than it exports. The national debt is the result of the fact that the US government spends more revenue than it collects. There’s no direct relationship between the two.
The whole article is written to claim that there is no connection. But anyone who knows the sectoral balances identity will recognize that there is a connection.
So the sectoral balances identity is:
NL = DEF + CAB
where, NL is the private sector net lending, DEF is the government’s deficit and CAB is the current account balance of payments (and is to a zeroth order approximation, exports less imports).
Of course, this is an identity and shouldn’t be confused for any behavioural hypothesis but it’s still useful in creating a narrative around a model (such as an SFC model) with behaviour for households, firms, the financial system, the government and the rest of the world.
On an average the private sector net lending is a small positive number relative to GDP (such as 2%), although it can fluctuate a lot. So for the U.S. economy, we saw that it was negative a lot in the 2000s and then reverted to a large positive just before and during the crisis.
One should of course keep in mind the correct causality connecting the three terms. What connects them is demand and output at home and abroad.
So the government’s deficit is affected by exports and imports. If there’s a large current account deficit, there’s a fall in demand and hence output and hence income and hence taxes leading to a higher government deficit than otherwise. Deficits in turn affects the public debt.
In other words, the U.S. public debt would have been lower, ceteris paribus, had the problem of the U.S. trade imbalance would have been addressed. This would have happened because of higher national income and higher taxes.
People don’t see the connection because they are comparing different nations. So for example, Japan has been successful in international trade and yet has a high public debt. In the other extreme, Australia has had large current account deficits over the years and public debt much lower than Japan (relative to GDP). But one should do a ceteris paribus comparison.
See Part 2 here: Public Debt And Current Account Deficits, Part 2
FRED Graph: US 🇺🇸 Sectoral Balances
Tracking the sectoral balances of a nation’s economy is a great way to build a narrative about its economic dynamics. It’s true, that an accounting identity doesn’t say much about causation. But they hint it and if we have a behavioural model around it, then we learn a lot more. It was used by Wynne Godley to highlight the predicament on the horizon in the 2000s.
The Federal Reserve Bank of St. Louis has a nice website FRED where you can create charts and even observe them when new data gets updated. Below is the U.S. sectoral balances chart. This is a static image file. For the dynamic chart, track the link above (the header of this page).
The blue line is the private sector net lending—its income less expenditure— which was in deficit and led to the crisis. The red line is the government’s deficit and the green line is the current account balance of international payments.
There is no branch of economics in which there is a wider gap between orthodox doctrine and actual problems than in the theory of international trade.
– Joan Robinson, The Need For A Reconsideration Of The Theory Of International Trade, 1973
Orthodox trade theory tells us that the “market mechanism” should work to resolve imbalances in the current account of balance of international payments. Although, the economics profession has conceded that Keynesianism is correct, it is still far from thinking clearly about international trade.
So it is a bit surprising that The Economist would say something unorthodox about this. In a recent article it complains about Germany:
This is the Post-Keynesian idea that surplus economies put a burden on deficit economies.
A fiscal expansion by the German government has the effect of raising domestic demand and imports and reducing the German current account balance of payments. This allows the rest of the world to grow both because of German imports and also because they are less “balance-of-payments constraint”.
Second, Brad Setser has a blog post on the current account surplus of the Republic of Korea (South Korea).
It’s impressive to see Setser get the causality right:
Fiscal policy alone doesn’t determine the current account (even if tends to be the biggest factor in the IMF’s own model). A boom in domestic demand, for example, would improve the fiscal balance and lower the current account surplus, just as a fall in private demand improves the current account balance while raising the fiscal deficit.
The current account balance, government’s budget balance and the private sector financial balance are related by an identity and sum to zero. But the identity itself shouldn’t be confused with causation.
The correct causation between the balances is between domestic demand and output at home versus abroad. This causality has been highlighted by Wynne Godley in the past. See more on this blog post by me here.
Simon Wren-Lewis has an article on his blog on stock-flow consistent/coherent models by Wynne Godley. Unlike other articles, this has a more engaging tone and isn’t dismissive.
This is a good thing but it has the tone “Oh, there’s hardly anything new” about stock-flow consistent modeling and the sectoral balances approach. 🤦. To me this is highly inaccurate, to say the least. None of the models outside SFC models —with one exception—come anywhere close to the important question about what money is and how money is created. Even in the Post-Keynesian literature, while there are various non-mathematical approaches, there’s hardly anything that comes close. That important exception is the work of James Tobin as is summarized in his Nobel Prize lecture Money and Finance in the Macroeconomic Process. Except that Wynne Godley’s model greatly improve upon the deficiencies of Tobin’s approach.
The sectoral balances approach is a mini-version of stock-flow coherent modeling. Wren-Lewis seems to say there’s hardly anything great and don’t tell much. First, almost nobody was making a cri de coeur as much as Wynne Godley. Second, the approach makes it clear why a huge recession was coming. This is because US private expenditure was rising faster than private income and the US private sector was in deficit for long and the private sector was accumulating debt on a huge scale relative to income. It’s difficult to say when this would have reversed pre-2007, but had to reverse. Once this is reversed, i.e., when private expenditure slows relative to private income, so that the private sector goes into a surplus, output will fall as a result of a slowdown of private expenditure.
Moreover, the US economy had a critical imbalance in its trade with its current account balance of payments touching almost 6.5% at the end of 2005, hemorrhaging the circular flow of national income at a massive scale.
Wynne Godley’s argument was that because of the external imbalance, the US fiscal policy will be unable to expand output to full employment easily, once the US enters a recession. Hence, he proposed import controls for the United States.
None of anybody outside Wynne Godley’s circle came anywhere close to saying anything of this sort.
But these empirical analysis is a much more complicated discussion. At a simpler level, nobody has come closer to what stock-flow coherent models achieve. All we see is economists struggling with basic questions on how money is created, what role it plays and so on.
Wren-Lewis also criticises SFC models saying they have minimal behavioural hypothesis. Now, this is far from the truth. If you write stock-flow consistent models, which are more realistic, you’ll end up with having a lot of equations and parameters. Behaviour of each “sector” is articulated in these models. How money is created by the act of loan making by banks, to how households and firms accumulate assets and liabilities, to how firms making pricing decisions and how much they produce and how much households consume. In addition, the importance of fiscal policy is articulated: how governments make spending decisions, whether government expenditure can be thought of as exogenous and how in normal times—when politicians pay attention to how much the government’s deficit and debt it has—governement’s fiscal policy can be thought of as endogenous. And crucially, the supreme importance of the government’s finance in the financial assets/liabities creation process. While most economists stop at one time-step for the expenditure process, using stock-flow consistent models, you can see the full process. Moreover, the analysis highlights the correct direction of causalities. A good example is the direction of causation from prices to money.
I want to however highlight another important point. A lot about how the economy works can be understood without going too much into behaviour. Just national accounts, flow of funds and a minimal set of behavioural assumptions would be a great progress. The rest of the profession however struggles to even understand basic flow of funds. A lot can be understood because most of the times, economists are erring on basic accounting. Hence their story doesn’t add up and produces something completely unrelated to the real world. If only economists understood this, that’ll be a lot of progress. Stock-flow consistent models are rich in behavioral analysis but even without it, understanding flow of funds with a minimal set of assumptions is the right direction.
A well known economic identity states:
Snational = Inational + CAB
where Snational and Inational are national saving and national investment and CAB is the current account balance of international payments. In calculating national saving and investment, one adds saving and investment, respectively, of all resident sectors of the economy.
However, an accounting identity shouldn’t be confused with behavioural relationships.
Steven Roach is a good economist and it’s sad to see him confusing this. In a recent article for Project Syndicate titled America’s Trade Deficit Begins at Home, he uses this identity to conclude that if America wants to reduce her trade deficit, the solution is more saving.
Roach says:
What the candidates won’t tell the American people is that the trade deficit and the pressures it places on hard-pressed middle-class workers stem from problems made at home. In fact, the real reason the US has such a massive multilateral trade deficit is that Americans don’t save.
Total US saving – the sum total of the saving of families, businesses, and the government sector – amounted to just 2.6% of national income in the fourth quarter of 2015. That is a 0.6-percentage-point drop from a year earlier and less than half the 6.3% average that prevailed during the final three decades of the twentieth century.
Any basic economics course stresses the ironclad accounting identity that saving must equal investment at each and every point in time. Without saving, investing in the future is all but impossible.
A little thought on behavioural relationships tell a different story. The main causality connecting accounting identities is behaviour of demand and output at home and abroad. While it is true that by accounting identity, the U.S. current account balance will improve by more saving (such as households saving more, firms retaining higher earnings and government (both at the federal and state level) attempting to increase its saving tighten fiscal policy, it happens via a contraction of output.
Wynne Godley was one who stressed this before the crisis. In his paper The United States And Her Creditors: Can The Symbiosis Last? written with Dimitri Papadimitrou, Claudio Dos Santos and Gennaro Zezza, this is made clear:
A well-known accounting identity says that the current account balance is equal, by definition, to the gap between national saving and investment. (The current account balance is exports minus imports, plus net flows of certain types of cross-border income.) All too often, the conclusion is drawn that a current account deficit can be cured by raising national saving—and therefore that the government should cut its budget deficit. This conclusion is illegitimate, because any improvement in the current account balance would only come about if the fiscal restriction caused a recession. But in any case, the balance between saving and investment in the economy as a whole is not a satisfactory operational concept because it aggregates two sectors (government and private) that are separately motivated and behave in entirely different ways. We prefer to use the accounting identity (tautology) that divides the economy into three sectors rather than two—the current account balance, the general government’s budget deficit, and the private sector’s surplus of disposable income over expenditure (net saving)—as a tool to bring coherence to the discussion of strategic issues. It is hardly necessary to add that little or nothing can be learned from these financial balances measured ex post until we know a great deal more about what else has happened in the economy—in particular, how the level of output has changed
[boldening: mine]
This was pre-crisis from a few who were avowed Keynesians all their life! It’s unfortunate to see Steve Roach make an error even after so many years into the global economic and financial crisis. One should study Keynes seriously. While I am sure Roach appreciates the paradox of thrift, he forgets applying it to the analysis of United States of America’s trade deficits.
The new issue of ROKE is out and celebrates 80 years of The General Theory. Nick Rowe has a new paper in the issue titled, Keynesian Parables Of Thrift And Hoarding. Requires access. Nick has a post on his blog where he welcomes comments.
Abstract:
I argue that Keynes missed seeing the importance of the distinction between saving in the form of money (‘hoarding’) and saving in all other forms (‘thrift’). It is excessive hoarding, not excessive thrift, that causes recessions and the failure of Say’s law. The same failure to distinguish hoarding from thrift continues from The General Theory into the IS–LM model and into New Keynesian macroeconomics. On this particular question, economists should follow Silvio Gesell rather than John Maynard Keynes. The rate of interest in New Keynesian models should be interpreted as a negative Gesellian tax (that is, a subsidy) on holding money issued by the central bank.
In my opinion, a part of it, the distinction between what’s called “hoarding” and “thrift” above is not something which Keynesians haven’t considered. In fact, in the sectoral balances approach, it is net lending and not saving whose behaviour is more highlighted.
A sector’s or an economic unit’s saving is defined as its disposable income less consumption expenditure.
S = YD − C
On the other hand, a sector’s net lending is defined as its disposable income less expenditure.
NL = YD − C − I
These things are not as straightforward as they look. Here’s an example. I can be both a saver and borrower.
Let’s say, I start with no assets/liabilities, earn $1mn in a year, pay taxes of $200,000, have consumption expenditure of $100,000 and buy a house worth $5m by borrowing $4.3m from a bank.
My saving is $700,000 and my net lending is minus $4.3 million [ = ($1mn − $200,000) − $300,000 − $5m].
Of course, the fact that I am a net borrower doesn’t make me poor. My house is worth $5m and I have a liability of $4.3m which implies my net worth is $700,000.
However, my liquidity is low. If tomorrow the economy collapses and I lose my job, I will be in a bad situation. In short, negative net lending (or a negative financial balance) of an economic unit or a sector contributes to financial fragility.
There’s another important point: even though I am a saver, my saving rate is 7/8, I have contributed hugely to aggregate demand. This can happen at a sectoral level as well. So perhaps this is the reason why Nick Rowe makes this distinction.
So if we were to blindly believe in Keynes, we would have concluded wrongly by just looking at the saving rate. But it is a matter of emphasis: economists make ceteris paribus arguments and I do not think Keynes didn’t understand this. He was perhaps holding everything else constant and changing the propensity to consume to highlight an important fact. But in real life ceteris is never paribus. What Keynes was arguing was that saving is not necessarily a good thing at the macro level.
Back to sectoral balances. Since, a sector’s (such as the household sector’s) negative net lending (or negative financial balance) adds to its financial fragility, this process will reverse. Private expenditure relative to private disposable income will fall. But this has an effect of being a drain on aggregate demand. And this can cause a recession.
Nick Rowe would have argued that it is the demand for money which caused a recession. Till here, it’s the same as argued above, because private expenditure falling relative to income is due to economic units trying to increase their liquidity. (There’s a “paradox” here: all units trying to reduce their fragility causes more fragility!).
There is however a difference: a sector’s or economic units’ demand for “money” can also be independent to income/expenditure and is more related to asset allocation between various kinds of financial assets. But here it cannot be said to cause a fall in aggregate demand and output. So a higher demand for money per se cannot be said to cause a recession.
As I was finishing writing this, JKH put up a comment at Nick’s blog saying Keynes understood it. I reproduce the comment below.
JKH:
There’s no question that Keynes appreciated the distinction between thrift and hoarding:
GT Chapter 9
“The rise in the rate of interest might induce us to save more, if our incomes were unchanged. But if the higher rate of interest retards investment, our incomes will not, and cannot, be unchanged. They must necessarily fall, until the declining capacity to save has sufficiently offset the stimulus to save given by the higher rate of interest. The more virtuous we are, the more determinedly thrifty, the more obstinately orthodox in our national and personal finance, the more our incomes will have to fall when interest rises relatively to the marginal efficiency of capital. Obstinacy can bring only a penalty and no reward. For the result is inevitable.”
GT Chapter 13
“The concept of hoarding may be regarded as a first approximation to the concept of liquidity-preference. Indeed if we were to substitute ‘propensity to hoard’ for ‘hoarding’, it would come to substantially the same thing. But if we mean by ‘hoarding’ an actual increase in cash-holding, it is an incomplete idea — and seriously misleading if it causes us to think of ‘hoarding’ and ‘not-hoarding’ as simple alternatives. For the decision to hoard is not taken absolutely or without regard to the advantages offered for parting with liquidity; — it results from a balancing of advantages, and we have, therefore, to know what lies in the other scale. Moreover it is impossible for the actual amount of hoarding to change as a result of decisions on the part of the public, so long as we mean by ‘hoarding’ the actual holding of cash. For the amount of hoarding must be equal to the quantity of money (or — on some definitions — to the quantity of money minus what is required to satisfy the transactions-motive); and the quantity of money is not determined by the public. All that the propensity of the public towards hoarding can achieve is to determine the rate of interest at which the aggregate desire to hoard becomes equal to the available cash. The habit of overlooking the relation of the rate of interest to hoarding may be a part of the explanation why interest has been usually regarded as the reward of not-spending, whereas in fact it is the reward of not-hoarding.”
Being the supreme macro-accountant (the first one really), he would be totally in tune with the general stock/flow consistency theme of the post-Keynesians.
Hoarding is a stock/asset allocation of liquidity, interconnected with the determination of the interest rate, as he notes above. He correctly rejected the idea of the interest rate as being determined by an “equilibrium” of saving and investment. He maintained correctly that those two measures are continuously equivalent.
Recession dynamics are a flow phenomenon as he describes it, using reconciliation of income accounting at two different points in time.
The behavior of liquidity, hoarding, and the interest rate is stock behavior (including hoarding) within that saving flow dynamic (including thrift).
There are two ways in which the terminology “net lending” is used.
In national accounts, it is the difference between saving and investment for any economic unit or a sector. “S − I”. It is the financial surplus.
Bankers and central bankers use “net lending” a bit differently. Here there is “netting” when redemptions are netted. For example, suppose a bank lends 100 units in one period and for simplicity assume all 100 are redeemed. Also assume that it makes 110 units of loans in the next period. In this case, net lending is 10 units.
But these two shouldn’t be confused.
Steve Keen should perhaps understand that the devil is in the detail and if he is interested in making accounting models of the economy, he should improve his accounting.
In a recent Forbes piece, he says:
So if the private sector is to finance the government sector’s surplus, and if the economy is growing at the same time, then there has to be a net flow of new money created by the banking sector—part of which expands the non-banking public’s money stock, and part of which finances the government sector’s surplus. Therefore the banking sector has to “run a deficit”: new loans have to exceed loan repayments (plus interest payments on outstanding debt).
Call this net flow of new money NetLend.
[emphasis: mine]
In the scenario assumed, banks have a surplus because presumably their operations are profitable. In the absence of fixed capital formation by banks, their undistributed profits are their financial surplus. Banks are net lenders in the national accounting sense (and hence have a financial surplus not a deficit). They net lenders in bankers’ language because gross new loans exceed redemptions.
While it is not wrong redefine terminologies, Keen is doing a sectoral balance analysis where deficit/surplus has a different meaning.
In short, in a growing economy, if the government is in surplus, it is more likely that non-financial corporations and households together have a deficit or a negative financial balance. Gross new loans by banks exceeding gross redemptions does not imply banks have a deficit (i.e., a negative financial balance). It is of course not impossible for banks to have a deficit in such scenarios: consider for example, banks purchasing a lot of buildings for their offices. In that case, banks’ “S − I” may be negative.