Recent Posts

Wednesday, February 27, 2019

Primer: On The Inevitability Of Debt Finance

One needs to be careful about the effects of aggregation in models. One particular problem is the implications of the Kalecki Profit Equation. It suggests that private investment is self-financing, since investment is a source of profits. This is true in aggregate, but the reality is that the firms that increase investment are not necessarily the ones with increased profits. The financial system is required to bridge financing flows between such entities.

(This article is an unedited first draft a section in my book on recessions. It is missing a discussion of the observed data on debt growth. I expect that I will have a short potted history of recessions and the associated data in another section of the book, so I will not repeat the historical analysis in this section. The rapid growth in mortgage debt as associated with the housing bubbles ahead of 2008 are quite well known, and so I assume that readers will take my word for it. The text here is generic, and corporate debt growth was less marked than was the case of the household sector. I don't view this as a major conflict with what I say, as corporate fixed investment has been relatively weak since the 1990s.)

When we look at the economy, it is characterised by circular flows: one entity’s outgo is another entity’s income (using “income” loosely). This is why we can see upward spirals in nominal debt outstanding: there is no law of nature to stop it. The tendency for money outflows to return to the source entity because of these flows is known as reflux. For a large entity like a central bank or central government Treasury, these reflux effects are important, and need to be taken into account when discussing policy. However, for a private sector entity, the reflux effect is generally going to be small, and safely ignored. Even if we ignore the special nature of central government finances (as discussed in Understanding Government Finance), the importance of reflux flows means that government finance is not like household finance.

We can now examine a simple example to see this effect. For simplicity, we will just look at the flows between a few private sector entities, and leave out complications like inventories and dividends.

  • Assume that there are three firms, A, B, and C, and each pays its workers $100 in the accounting period to produce goods.
  • The workers spend all wage income, buying $150 of consumer goods from firms A and B. This gives each of the firms a profit of $50, which is $150 revenue less the wage bill used to produce the goods ($100).
  • Meanwhile, both A and B buy $100 worth of capital goods produced by firm C. The purchase of capital goods is a cash flow drain, but it is not an expense in the current accounting period. (The capital would be depreciated, which will be a non-cash expense in future accounting periods.) The net result of the cash flows for A and B is that each has an outflow of $50 in cash instruments.
  • Firm C sells $200 worth of capital goods, with a cost of production of $100, leaving a profit of $100. Since we assume that Firm C supplies its own investment goods, it has no investment outflows. The cash inflow matches the level of profits, or $100.

The aggregate profit for the business sector is $200 ($50 each by A and B, and $100 for C). This matches the cut down Kalecki Profit Equation: investment is $200, household savings is $0, and we have excluded all the terms from consideration. Although aggregate profits match aggregate investment, the firms doing the investing had total a cash outflow of $100. They could finance the investment by either transferring existing cash holdings, or more likely, issue debt instruments.
Since firms do not have an inexhaustible supply of cash instruments, the usual mode of financing is to issue debt. This could be done via vendor financing: firm C lends A and B the cash to buy the capital goods. In fact, this is typically how non-capital goods are financed; if we treat accounts payable as a form of debt instrument. The implication is that the non-financial sector can bypass the financial sector for financing needs, or to paraphrase Minsky: all entities act like banks. However, most fixed capital investments are too lumpy to safely financed via vendor financing (as the technology sector re-discovered the hard way in the late 1990s), and so they are normally financed by long-term debt contracts intermediated by financial firms.

Of course, large, profitable firms can self-finance. To the extent that the economy is dominated by established conglomerates, this would reduce the need for external debt financing. Nevertheless, the tendency is for more speculative investment to be undertaken by rising firms that are challenging the existing economic structure.

The tendency for investment to be financed by challenger firms also raises issues with aggregate debt statistics. If profitable established firms are paying down debt while new firms are borrowing large amounts to finance speculative profits, the riskiness of business sector debt will be rising even if the aggregate amount of debt outstanding is falling. Analysts would need access to micro-data – such as tracking the universe of debt issuers – to determine whether the risk profile of debt is changing.

From a modelling perspective, the breakdown in aggregation is a problem for approaches that are based on behavioural approaches to the aggregate, such as assuming that profits are maximised. For example, if we assumed that the business sector was in aggregate maximising profits (somehow), the aggregate behaviour would take advantage of the self-financed nature of investment. Stock-flow consistent (SFC) models deal with this by using heuristics that are based on average behaviour. Agent-based models are disaggregated, and so can capture the range of possible outcomes. However, the difficulty with agent-based models is fitting them to data, which is aggregated.

Equity financing is an alternative to debt financing, and so it might appear possible to break the link between investment growth and debt growth. However, the use of equity financing is limited in modern practice, for a number of reasons.

  • The premise of equity markets is that investors are targeting a common (risk-adjusted) rate of return on equity. However, the rate of return on assets is much more variable, and lower than this target. The use of debt financing allows firms to equalise (risk-adjusted) rates of return via magnifying the return on assets via leverage. Since interest payments are taxed and dividends are not, the tax system helps magnify the effect of leverage.
  • Debt instruments have finite terms, allowing firms to match financing to the economic lifespan of assets.
  • Housing is generally financed by mortgages. It is unclear what entity would be willing to become a junior partner in the equity of homes, nor would homeowners welcome interference from such a partner.
  • The credit markets (including the banking sector) have the capacity to match up the portfolio allocation desires of the private sector and debt issuance. The private sector is trapped in a web of nominal obligations (household bills, wage contracts) even if their balance sheet is clean; they want to hold cash instruments to match those outflows. Equity has a highly uncertain value, and so cannot be the entirety of private sector portfolios.

It is possible that equity will be a more important source of finance in the future. However, this would require changing attitudes within society, such as somehow lowering the target rate of return on equity. Until then, debt will tend to be correlated with investment.

We can now turn to the role of the financial system. Many popular discussions of finance assume that the role of the financial system is to increase the stock of saving to raise investment. This is a flawed application of loanable funds thinking. The role of the financial system is to allocate financial flows: match savers to borrowers, and have pricing on loans properly matched against risks.

We need only look at the well-documented disastrous practices during the Financial Crisis to see that the financial system is perfectly capable of failing to allocate flows properly. The post-Keynesian argument is that practices in the financial system are a far more likely cause of (serious) recessions than changes in the real economy, as will be discussed in Chapter TK.*

Nevertheless, so long as the financial system is operating in something resembling a normal fashion, borrowers will be able to finance investment at reasonable risk-adjusted rates of interest. The prospects for returns on capital – real economy analysis – can still drive the business cycle. Debt issuance will just move in line with financing requirements, and so be a determined factor, not a determining factor for growth.


* Why "Chapter TK"? Old writing trick. The letter combination "tk" is extremely rare in the English language (unless one is writing the memorabilia of Dick Butkus). It is very easy to search for placehlders if you use that combination. The combination "xx" is also popular - until you start writing about the Euro Stoxx(tm) index...

(c) Brian Romanchuk 2019


  1. I would like to offer an alternative, realizable model. First we build the basic model, then introduce debit.

    This model has one producer who builds and sells a product, workers who trade labor for money, and a beginning supply of money. The money supply is initially limited but, as we shall see, full output is still possible.

    We will take advantage of the observation that workers usually get paid after working a period of time (one week in this example).

    Initially assume that the entire money supply is in the ownership of producers. Workers can not buy because they have no money. This is not a problem because workers get paid once a week.

    Production begins by hiring all workers, expecting to have funds for one week's pay. At the end of one week, production completes and is completely distributed to outlets, ready for sale. Workers are paid the entire money supply. The second week has production idle (due to lack of money) while workers spend all their earnings to consume last weeks production. At second week's end, workers are out of money but the entire money supply is back in producers ownership.

    This cycle is repeatable and realizable.

    This model demonstrates inefficient use of resources . It can be improved by increasing the frequency of exchanges and increasing the number of pay periods (not only once a week).


    This model has a severe weakness. If workers fail to completely return money (by buying product) the producer would be unable to hire the entire workforce the second period. This could be cured if the worker would take debt in exchange of work OR if the worker loaned money to the producer.

    Logically forecasting, the producer would be foolish to allow the worker to lend him money IF THE MONEY WAS OBTAINED BY NOT BUYING PRODUCT. On the other hand, the producer would be a little more willing to borrow if the worker agreed to take a claim on production so that the producer still knows the product will be consumed (probably). This would be akin to a worker taking a limited gift certificate in exchange for work.

    Now we have an alternative, realizable model with all the elements you included in your model (except profit).

    Unlike your model, we don't yet show profit. Instead, we show a bump in the flow of money that is mitigated by re-establishing flow with debt. The wisdom of accepting the mitigating method is up to both parties (to the debt exchange).

    There you have it -- an alternative model.

    1. Hello, I'm in the middle of writing a complex article, and I can't be distracted when I'm doing such work. However, I think you are missing my objective: describe an observed empirical regularity, not potential alternatives. By using suitable assumptions, there are an infinite number of behaviours that are consistent with accounting identities -- which are the only "laws of nature" that must apply. However, in the observed data, we do see particular patterns, that need to be explained.

    2. This comment has been removed by a blog administrator.

  2. This comment has been removed by a blog administrator.

  3. This comment has been removed by a blog administrator.

  4. This comment has been removed by a blog administrator.

  5. This comment has been removed by a blog administrator.

  6. This comment has been removed by a blog administrator.

  7. This comment has been removed by a blog administrator.

  8. This comment has been removed by a blog administrator.

  9. This comment has been removed by a blog administrator.

  10. This comment has been removed by a blog administrator.

  11. This comment has been removed by a blog administrator.

  12. This comment has been removed by a blog administrator.

  13. This comment has been removed by a blog administrator.

  14. This comment has been removed by a blog administrator.

  15. This comment has been removed by a blog administrator.

  16. This comment has been removed by a blog administrator.

  17. This comment has been removed by a blog administrator.

  18. This comment has been removed by a blog administrator.


Note: Posts are manually moderated, with a varying delay. Some disappear.

The comment section here is largely dead. My Substack or Twitter are better places to have a conversation.

Given that this is largely a backup way to reach me, I am going to reject posts that annoy me. Please post lengthy essays elsewhere.