Not a “Savings Glut” per se but a monetary excess amidst a period of complex global structural economic change!

If you stream through the data it is pretty clear that developed economy growth has been slowing for some time and that monetary policy has accommodated this adjustment with lower interest rates and a relaxed attitude towards money supply growth.  At about the same time these trends were moving ever closer to their sweet spot on the horizon (because we are not yet at peak of this particular movement) certain developing markets really got going, with the help of a fair amount of their own monetary stimulus but also by a reconfiguration of global supply chains and offshoring in key economies.  All factors combined to create a heady and dangerous global financial imbalance, a weak bridge cast across a widening economic divide.  No wonder it all came crashing down..but who was to blame?  The world’s central bankers who were blindsided into excessively lax monetary policy by a low inflationary world that had become obsessed with laying off and chopping and dicing of risk to those “who could best absorb and bear it”, or some of the finer strings in the mesh?  Well, some have chosen to blame excess savings in the emerging/developing part of the world, principally China, but this is all too pat.   The “savings glut” theory, if you can really call it “excess savings”, was merely a return of serve of part of the vast ocean of financial and monetary excess that barrelled through the early to mid 2000s.

For some the current global crisis is the direct result of “excess savings”: countries that “save” more than they “invest”, invest their “excess savings” abroad, thereby apparently providing the fuel for consumers in those countries to borrow to spend on their goods and other expenditure. This surge of “savings” from overseas pushes interest rates down making it easier for consumers in those countries to borrow those “excess savings” and spend. The flow of “savings is = to the trade deficit”.

If you were to just look at the simple National Income Accounting identity, you may be taken in by these arguments: output in this identity is a function of consumption+investment expenditures with saving being output not consumed but still spent. Output being the sum of all expenditures is likewise equivalent to income, so saving =Investment.  BUT, this simple model does not incorporate expenditure that does not come from prior period income, in other words it does not allow for new bank loan generated demand.  

When a bank lends money it creates new deposits; these deposits are not only spent but also add to the money supply impacting the demand for assets and consumption/investment goods across the economy.  To ignore this source of demand would appear foolish, but the basic fundamentals of the “savings glut” argument do just that. 

The following chart shows the compound growth rate of broad US money supply growth relative to GDP growth, and you can see that the 15 year period to which Ben Bernanke refers in his 2005 “The Global Saving Glut and the U.S. Current Account Deficit corresponds closely with this increase in money supply growth relative to GDP.


A rise in money supply growth over and above the rate of growth in gross domestic product implies that more and more domestic money supply growth is becoming asset focussed, meaning that a strong underlying supply of domestically generated funds was available.  Indeed we have seen a large increase in household asset values relative to GDP….


Yes foreign exchange demand for US dollars and thence US assets would elevate the asset based velocity of domestically generated money supply and is one of the reasons why excess MS growth itself did not generate inflationary problems amongst consumption and investment expenditure items. 

So let us look at some excerpts from the 2005 Ben Bernanke statement in 2005:

“over the past decade a combination of diverse forces has created a significant increase in the global supply of saving…which helps to explain both the increase in the U.S. current account deficit and the relatively low level of long-term real interest rates in the world today”

“When U.S. receipts from its sales of exports and other current payments are insufficient to cover the cost of U.S. imports and other payments to foreigners, U.S. households, firms, and governments on net must borrow the difference on international capital markets”

“Thus, essentially by definition, in each period U.S. net foreign borrowing equals the U.S. current account deficit, which in turn is closely linked to the imbalance in U.S. international trade.”

This…perspective focuses on international financial flows and the basic fact that, within each country, saving and investment need not be equal in each period.

All investment in new capital goods must be financed in some manner. In a closed economy without trade or international capital flows, the funding for investment would be provided entirely by the country’s national saving.

If a country’s saving exceeds its investment during a particular year, the difference represents excess saving that can be lent on international capital markets. By the same token, if a country’s saving is less than the amount required to finance domestic investment, the country can close the gap by borrowing from abroad

As I have discussed today, the underlying sources of the U.S. current account deficit appear to be medium-term or even long-term in nature, suggesting that the situation will eventually begin to improve, although a return to approximate balance may take some time. Fundamentally, I see no reason why the whole process should not proceed smoothly”

Bernanke says “in a closed economy, the funding for investment would be provided entirely by the country’s national savings”. This key assumption is drawn from the simple national income identity (Output=C+I) and national savings is Output minus Consumption..S (I)=Y-C.

The problem is this ignores the role of money supply growth: the only way (apart from QE) that money supply grows is via bank loans which create new deposits. The national income accounting identity is also a static model and cannot deal with how economies grow and finance their growth from period to period.

In fact, income received from economic activity today is a function of income received in the prior period + new loan/money supply growth. Any outcome that we could define as “excess saving” therefore has its roots in something other than the national income accounting identity.  Relying on the identity divorces our analysis from excesses and other structural issues that are critical to correctly analysing economic, financial and market relationships.  I will explore the NIA identity in a separate post.

What we are essentially trying to define is the stability of the demand flow on the one hand and the fund flows on the other, and in the case of the “savings glut” argument, the root of the funding and the direction of causation.

Attributing the financial and economic excesses that led up to the 2007/2009 crisis (and beyond if innovative monetary policy is a guide) to a mere savings imbalance overlooks the foundations of financial market risk and the divergence between the asset/financial side and the real economy.   We had latitude (markets and finance) and longitude (economic) problems. 

To go back to Ben Bernanke’s argument:

“When U.S. receipts from its sales of exports and other current payments are insufficient to cover the cost of U.S. imports and other payments to foreigners, U.S. households, firms, and governments on net must borrow the difference on international capital markets”

“Thus, essentially by definition, in each period U.S. net foreign borrowing equals the U.S. current account deficit, which in turn is closely linked to the imbalance in U.S. international trade

Well first of all we know US consumers financed much of their expenditure from new money loans, many of which used housing collateral as security: they did not borrow directly from overseas’ “excess savings”.  I will address the detailed mechanics of this in a later post. 

The argument that excess savings created the trade deficit is an easy one to accept, if you have not run through exactly what you are saying.  In the static one period NIA identity without new money creation where one country exports and the other imports, you need country 1 to increase consumption at the same time as country 2 cuts consumption for borrowing and excess saving to occur at the same time.  But in the real economy, it is hard to visualise how an NIA identity driven interaction of this nature could actually occur. 

The chart below shows net annual dollar flows into US assets (foreign flows into US assets net of US flows into foreign assets) and mirrors (though not precisely) the US trade deficit in dollar terms.


If you are importing goods you have to pay for them in the currency of the country you are buying from: you have to exchange dollars for the foreign currency.

If you are importing more than you are exporting the foreign country is going to accumulate your currency. It has a choice: sell it back into the FX market (forcing say the $ to fall and the Yuan to appreciate), or to use it to buy assets in that currency, effectively recycling the money back into the importing economy and supporting the export generating ability of your currency.   The dollar asset demand generated from this wheel is not a natural one but a forced one.

In reality this build up of trade accumulated foreign exchange, in China’s case, has largely been bought up by Central banks though issuing their own currency in exchange for dollars. This is hardly saving, but it is funding.  Central banks then use the $ currency acquired to buy US dollar assets. In reality they are using money newly created by a central bank to buy US money supply previously created by the US financial system to purchase assets so that their own currencies can remain competitive. I will look further into the NI identity with respect to this process and its relevance to saving in a later post, but for the moment the use of newly created CB money is a funding flow that is exchanged for money created by the US financial system thereby increasing the asset focus of the US money supply.  Chinese resident exporters receive their funds in Yuan, Yuan which feeds into the true Chinese inter temporal national income accounting identity, which is both a separate and a potential complementary dynamic to that provided by Central Bank activity – this is a complex area which I would like to address at a later date since CB action creates new money   

The problem is ascribing the blame behind the trade deficit squarely at the root of “excess savings” when the actual savings transaction post the FX exchange has likely yet to occur. In reality much of the flow of capital back to the US is more like a return of serve than a new service game. And, it is important to bear in mind the sequence of flows(you have to import goods from abroad before the foreign country is able to accumulate FX for purchase) and you have to bear in mind the original source of the flows (in this case a growing broad money supply supported by new bank loans). 

Yes, recycling dollars into US denominated asset markets is going to impact the portfolio demand for assets, it will raise prices and will cause yields to fall below where they may otherwise have been but a) these are only net flows, b) they would not have impacted the ability of the Federal Reserve to set the Federal Funds rate, and c) they do not imply causation.

So, let us look at the trade deficit, net foreign purchases of US assets, US broad money supply growth and changes in non financial sector credit market debt. Well we can see that money supply growth on its own is significant and should be considered alongside net foreign investment flows as part of the equation lying behind excessive consumption, debt expansion and increasing asset values in the US during this period.


We can look further at the cumulative increase of broad money supply growth, the trade deficit and net flows to US assets:


Cumulative growth in broad money supply growth significantly exceeds the cumulative growth in the net trade deficit/net foreign asset purchases.  In point of fact, the flows are not necessarily comparable because money is a stock of which transactions are a multiple, so the significance of money is of a higher order. Correctly, many papers (Shin 2011, Borio2012 etc) recommend looking at gross financing flows as opposed to the narrower trade deficit.

And as noted in the first chart above, if we look at a wider historical frame we can see that broad US money supply growth has reached an unprecedented level with respect to nominal GDP growth, so the static Y=C+I(S)+X-M model is ignoring significant shifts in domestic and global financial stability.

We can also look at US new consumer credit and home equity line of credit growth relative to Personal Consumption Expenditure and disposable income:


At one level we can see that consumption growth significantly exceeded basic wage growth, at another we can see that the availability of new credit to finance consumption was significant.

In terms of financing personal consumption it is difficult to see a gap of any real significance that would need to be funded from abroad.  Nominal $ savings levels were also palpably stable to end of 2004, at least (see below), although we did have a noticeably lower level of real growth in personal disposable income per capita and a debt financed property boom over this period.  One of the reasons savings as a proportion of income probably fell during the period was possibly because of the higher level of debt financed consumption expenditure, given that expenditure numbers do not differentiate between income driven and loan driven financing of consumption.  Again, inter temporal dynamics may well obscure the ability of simple identities to differentiate the many threads comprising economic activity.


It is also important to note that real income growth was lower on an historical basis over this period relative to other post war growth cycles:


Importantly, as noted in a number of papers (Shin, Borio and Disyatat) European financial institutions were more heavily embroiled in mortgage debt securitisation than the supposed trade deficit “excess savings” band of emerging/developing economies.   While many of these papers themselves are unclear as to what money creation model they use, the flows and their sources are relevant. We can see below a chart that shows cumulative household mortgage debt relative to cumulative growth in commercial bank liabilities (a proxy for broad MS growth): note the significant change in this differential post 2001, indicative I believe of high levels of securitisation.


Another area of importance was the development of global supply chains, especially centred in emerging Asia, and a number of key global trade deals.  Again this is something discussed in a number of papers criticising the Savings Glut hypothesis (Palley) and others (Pierce and Schott). 

With a lot of manufacturing moving offshore it is not surprising that imports also surged: indeed we can see from the following chart that the value of US net direct overseas investment (net of foreign direct inward investment) relative to the US trade deficit was in substantial surplus around this time. 


As opposed to capital coming into the US to fund direct investment, there appeared to be more capital flowing the other way, confirming perhaps the significance of the impact of offshoring.


Indeed we can also see US net investment income also appeared to respond significantly during this period.   Worth noting that if the rise is driven by profits on overseas operations related to offshoring, then we are talking to some extent of a large savings component on US import flows that would effectively be considered US exports and not foreign savings, even though it would appear that offshore profits are being increasingly held overseas.  



Other extraneous data add to the chinks in the excess overseas savings theory as the prime cause of the financial, economic and market crisis that engulfed the world in 2007 and beyond include the incredibly large amount of share buybacks.  For instance when tracking share buybacks as a % of GDP there is most definitely a visually interesting reflection of the trade deficit.  To me this smacks of accommodative monetary conditions, for both sets of data, as opposed to an overwhelming urge on behalf of the US’s trade deficit partners to send hoards of savings to the US.


We have also seen large overseas profits of US companies accumulate in offshore bank accounts.  To what extent were these profits sourced from US import demand?  This is relevant where US dollar flows associated with these imports have already been absorbed by overseas central banks and reinvested in assets. 

In summary we have had a number of different dynamics interacting to produce the crisis, but it is certainly a lot more complex than a “savings glut”.

A weakening in the developed world’s growth rate led to monetary policy accommodation and a surge in asset focussed money supply growth and debt that goosed consumption and created unstable imbalanced growth.  The developing world had its own rate of development accelerated, but again likewise imbalanced and increasingly indebted. 

It was never going to be possible to keep developed economy growth at the higher levels we had been used to in the post war period, while temporarily boosting it through monetary means, in the hope that developing economies could mature fast enough to allow slowing growth in one and growth potential in the other to meet up before monetary accommodation exhausted itself.    

There was not a “savings glut” per se, but there was an excess supply of global demand for financial assets and an excess of consumption in parts less able to afford it.   Clearly it is a heavily nuanced and fuzzy frame complicated by intervention that has, rather than solved the problem, placed greater strain on its structural integrity.   As the world’s economic dynamics change, sometimes for the worse and sometimes for the better, we need to be really clear as to what impact our actions are likely to have.   We could have easily had lower economic growth, lesser debt and lower asset prices, but on the plus side we would most likely have far less to fall because the current divide between asset prices and underlying economic fundamentals are unparalleled.

The economic frame of the last decade or more has been one with different directional dynamics, but the powers that be decided to juice it as if it were one, at a critical time in its transformation, as if the global economy was merely something that needed to be prodded heavily enough into life.   

Leave a Reply