Recent Posts

Wednesday, October 15, 2014

Using Stock-Flow Norms To Explain Secular Stagnation (Part 2)

Having looked at the increasing demand for financial assets in the first part of this essay, we now turn to look at the supply of financial assets. The business sector is unable to supply the required amount of financial assets, leaving balancing supply and demand to the government and household sectors. Once the housing bubble burst, the government sector was put into the position of being the key source of supply. This can only be accomplished by large fiscal deficits, which implies slow growth. Since this slow growth was the result of policies that were aimed at increasing growth, it is unlikely that this structural situation will be easily changed.

The reason I emphasise the role of stock-flow norms is that the levels of savings flows are not remarkable at present. What is unusual at present is the high ratio of the stock of financial asset holdings relative to income (in this case, I scale by domestic income - GDP). The desire and ability of investors to hold a large stock of assets - documented in Part 1 - forces the stock of supply higher. This is what sets the present era apart from the rest of the post-war period.

Business And Foreign Sector Supply Of Financial Assets

Chart: U.S. Financing Gap

One non-obvious regularity that we see within the national accounts is that business sector investment is largely self-funded. The chart above shows the financing gap for U.S. Nonfinancial corporate businesses. It is defined as (from the U.S. Flow of Funds report):

Capital expenditures less the sum of U.S. internal funds and inventory valuation adjustment.
What we see is that the financing gap is relatively small (from a macro perspective) - often on the order of 1% of GDP. What does this tell us? It essentially says that internally generated funds are roughly enough to finance capital expenditures. (This is not too surprising from the perspective of the Kalecki profit equation.) The amount of borrowing they undertake may deviate from the financing gap, as shown below. In the current environment, there is more borrowing than is indicated by the financing gap; this is likely the result of borrowing to fund dividends and share buybacks. However, this use of funds is just borrowing from one set of investors to hand money to any set of investors, and so it just a reshuffling of the financial assets available.

Chart: Nonfinancial Corporate Net Borrowing

(As an side, we exclude the financial sector as it just borrows to fund positions in loans that are ultimately taken out by the nonfinancial sector, other than those needed for the financial sector's relatively small capital expenditure needs.)

Examination of the above charts show that the current environment (post-2001) is different - the financing gap dropped to low levels, and was negative for most of the post-crisis period. This means that the corporate sector does not need to borrow to invest, and so the desire to increase financial asset holdings outside the business sector is not finding an outlet in increased fixed investment. For example, as I showed in Part 1, the retirement savings complex is steadily purchasing credit market assets that amount to a flow of about 1.5% of GDP.
Chart: U.S. Net Equity Issuance
In this analysis, I focus on debt flows, even though equities are a large portion of private sector portfolios. This is because net equity issuance has turned negative since the mid-1980s. The equity market is no longer a net source of funding, its role is to allow owners of private companies to liquefy their positions, and to send pricing signals. Market capitalisation is only able to grow as a result of upward revaluation of shares, which is aided by the reduction of shares outstanding generated by buybacks. This generates instability, as valuations are more stretched than in earlier eras. Moreover, it means that private savings flows - which includes the proceeds of sold equities - have to be invested in fixed income markets.

The financing gap was quite low in the early 1950s as well, but growth was buoyant. The difference is that the United States has a current account surplus during that period, which means that the United States was exporting capital. (The French used to complain about the United States running trade deficits, but those trade deficits were actually financed by income flows - the current account was in surplus. It was not until the Reagan era that the United States really exploited whatever privileges "reserve currency" status gives it.)  The United States is now importing financial capital - which gives rise to the explosive foreign financial asset holdings which I showed in the previous part - and those savings have to find a home in the domestic economy.

The Household Sector's Attempt To Supply Financial Assets

Chart: Mortgages Outstanding

The chart above illustrates the U.S. household sector's doomed attempt to supply financial assets - in the form of mortgages. It was unable to service the debt, and the sector has been forced to retrench. At present, there is increased issuance of dubious auto loans and student debt, but these debts are obviously low quality, and so they are unlikely to provide a large enough outlet for inflows to financial assets.

The Government - Last Sector Standing

Chart: Treasury Securities Outstanding

Since the other sectors are unable to provide the financial assets to allow entities to hit their target wealth-to-income ratios, the government has to fill that role by default. Returning to the Treasury -to-GDP ratio chart above, its movements make much more sense in the context of the previous analysis. The ratio fell from the high levels from after World War II to hit a low in the inflationary 1970s. It then reversed higher, which is the result of disinflation and the increased desire of entities to hold a higher ratio of assets to income.
Chart: U.S. Treasury And Mortgages Outstanding

The Clinton-era reduction of the debt-to-GDP ratio was only a temporary reprieve from this trend. Mortgages moved to replace Treasurys within private portfolios. As shown above, the aggregate of these markets followed a smoother trend, which suggests that the driving force for the dynamics was investor preferences. Once those mortgages proved to be bad assets, Treasury securities reverted back to a more prominent position within investor portfolios.

Whither Secular Stagnation?

Although the ratio of financial assets-to-GDP may stop rising, it has reached such a high level that just keeping the ratio steady implies much more debt issuance than the corporate sector can safely sustain. The imbalance between desired savings and investment is helping cap nominal GDP growth at the 4% annual rate we have seen this cycle. Although the economy is not in a deflationary state, this is not enough to reduce the significant slack that remains in the labour market, as evidenced by various measures of under-employment.

"Fixing The Debt" - Not Any Time Soon

(I should preface this section by noting that I believe that the government debt-to-GDP ratio is a piece of meaningless trivia for nations with a free-floating currency. However, not everyone agrees with that assessment, and so I will discuss whether it can be easily lowered.)

Although the debt-to-GDP ratio will trend lower until the next recession hits (and depending on the current turmoil, that could be soon), it will take some major policy changes to reverse course.
  • Increasing wage inflation will increase the pace of GDP growth, lower the profit share of GDP, and increase discount rates. This will have the effect of lowering financial asset-to-GDP ratios, and hence debt-to-GDP ratios. This is why sensible bond investors fear low debt-to-GDP ratios, not high ones. But to get this outcome, inflation-targeting would have to be dropped, and labour market reforms reversed.
  • The unwillingness of mainstream political parties to increase pay-as-you-go pensions, means that savings are going to be continuously thrown at the pension-finance complex. The replacement of defined benefit plans by defined contribution plans means that more savings are necessary, as defined benefit plans can take advantage of early mortality to reduce the amount needed to save for the same amount of post-retirement benefits. Although demographics may reduce the amount of retirement saving, there is little indication that retirement plans are doing well enough to allow that to happen.
  • The asymmetry of Fed policy - cut rapidly, and hike slowly - has driven nominal and real interest rates to low levels. This forces entities to accumulate a greater stock of bonds in order to achieve the same level of retirement income. To reverse this trend, the Fed might have to flip the signs on its models and hike rates in order to get the economy to speed up. This provocative analysis is one of Warren Mosler's most interesting insights.*
    The difference between now and the negative real rate period of the 1970s is that nominal rates were still quite positive, and so bond portfolios did generate reasonable income. Additionally, bond markets were in long-lived bear trend then, and so there were few investors who wanted to increase their bond positions. (Of course, a great many investors dislike the low yields available on Treasury bonds now, but bonds are still in a secular bull market, producing capital gains.) 
  • Although there are grumblings along the trade front, it is very hard to see globalisation being reversed. It appears impossible to reverse the direction of the United States' current account rapidly, even if which countries are running the surpluses changes. The belief that the U.S. trade deficit in oil will disappear is a pipe dream - roughly half is imported, and any oil price weakness will only induce consumers to buy even bigger vehicles.
  • Reducing inequality would probably lower the financial asset-to-GDP ratios, but there is no plausible reason to expect that to happen unless wage inflation rises. The idea behind this assertion is that it appears that the wealthiest segments of society have the greatest tendency to accumulate assets.
  • The corporate sector does not show any ability to ramp up investment without it resulting in a bubble, causing a corresponding crash when it pops. Therefore, we cannot expect corporate debt issuance to take over from government issuance in a sustainable fashion.
  • Shrinking the government's footprint in GDP may allow for a smaller government debt-to-GDP ratio, but it means that the private sector has to increase its debt issuance to make up the slack. This could possibly make a difference during an expansion, but it will probably reverse when the next recession hits. And with a smaller government, the post-recession recovery will be even slower, as private sector balance sheets would be more fragile.
As we see, pretty well any mechanism to reduce debt will require reversal of other signature policies of the recent decades. Although possible, I see no sign of mainstream political parties moving towards enacting such policies, as I am unaware of any politicians that would want to run a campaign with a platform calling for reduced wealth.


* There historically was a debate about whether higher interest rates would be inflationary, up until the 1970s. (Amongst market analysts, at least. I ran into that idea a few times when I went through the archives during my first financial job. I am unsure what the academic community thought about that topic at that time.) The Volcker episode moved the consensus firmly towards its present stance - higher interest rates reduce growth and hence inflation. I have some links and further discussion on this topic here.

See Also:
  • The relationship between bond yields, nominal GDP growth, and debt-to-GDP ratios.
  • Circuit at Fictional Reserve Barking quotes from a article by James Tobin from 1963 on this topic - Deficit, Deficit, Who's got the Deficit. The article also discusses the current situation. James Tobin was part of a group at Yale University which developed a body of stock-flow consistent economic modelling. When compared to the school of thought from Cambridge, associated with Wynne Godley (which I am more familiar with), the Yale researchers were more focussed upon portfolio construction and how it interacts with the economy.
(c) Brian Romanchuk 2014


  1. " the government has to fill that role by default"

    Perhaps not the best choice of words. :)

  2. Ouch! I was not a credit analyst, and had the D-word removed from my vocabulary...

  3. It would seem to me that the main problem with everyone's views on the economic impacts of interest changes is that they are not dynamic.

    It can never be as simple as "interest rate increases slow economic growth" or "interest rate increases speed up economic growth", because the relative impacts of interest rate changes are dependent on the size of each type of debt and the maturity structure (how fast interest rate changes impact interest income and spending).

    The fiscal impact of a 5% increase in the FFR cannot be determined without acknowledging the size and the maturity structure of outstanding TSY securities. If T-securities outstanding are only 50% of GDP, the fiscal impact would be much lower than at 250% of GDP.

    Shorter-term maturity structures on T-securities mean interest rate changes impact the interest spending channel quickly. If we had nothing but 30yr T-bonds outstanding, it would take along time for interest rate changes to impact interest spending by the federal Govt.

    1. The more I think about how interest rates are supposed to influence the economy, the less sense it makes to me. The standard story is that it should affect private sector savings/consumption decisions, and it should not be driven by the interest income channel. Therefore, the mainstream does not pay attention to the maturity structure of debt outstanding, other than looking at what the Fed owns.


Note: Posts may be moderated, and there may be a considerable delay before they appear.

Although I welcome people who disagree with me, please be civil.

Please note that my spam comment filter appears to dislike long "anonymous" posts. I get no warning about this, and only go through my "spambox" infrequently. The best bet it to keep comments short, and if you think the spam filter struck, let me know with a short comment.