...against fictions and other tall tales

Monday, 26 March 2012

Careful with the saving terminology: a reply to JKH

By Joseph Laliberté and circuit. The authors reside in Canada and come from two different heterodox backgrounds, namely, MMT and the post-Keynesian/circuitist tradition.


Much discussion took place lately regarding the proper meaning of the term “saving”. The issue surfaced when supporters of “Modern Monetary Realism” (a MMT offshoot) called out supporters of Modern Monetary Theory over the confusion they sometimes entertain when defining this term. Specifically, MMR contends that some MMTers often confuse “saving” with “saving minus investment” or “S-I”. The MMRers stress, correctly in our view, that “S” corresponds to “saving” rather than “S-I”. The anonymous blogger JKH has written extensively about this aspect and has even authored a detailed paper on this issue. We highly recommend this paper, as it provides an in-depth analysis and summary of this whole issue.

That being said, we wish to start off by saying that, while one could blame MMTers for putting too strong an emphasis on “S-I” rather than “S” in their analytical framework, it would be wrong to suggest that MMT’s lead proponents (i.e., Fullwiler, Mosler, Wray, etc) are not aware of the intricacies of the “saving” definition. Also, although it is true that some MMTers may have been sloppy at times in their use of terminology (for instance, we know of examples where MMTers have incorrectly referred to “S-I” as “saving” or “net losses”), it is unfair to characterize MMTers as being the only ones who get the terminology wrong. As Scott Fullwiler repeatedly and correctly pointed out, most of us at one time or another are guilty of using unclear or incorrect terminology, especially when commenting on blogs.

The objective of this post is to clarify terminology and demonstrate that JKH’s own choice of words regarding the notion of “saving” is not entirely without fault and could, as a result, mislead some readers. But before doing so, we wish to stress that we fully support the goal of both MMT and MMR (and JKH) in trying to put forth a more technically-sound approach to macroeconomics and economic policymaking. Indeed, it is our hope that all commentators would settle on a standard set of terminology moving forward.

Net Saving or Net Lending?

It is well-known that MMR and MMT have more or less officially adopted Wynne Godley’s terminology. Accordingly, both MMT and MMR would contend that “S-I” should be identified as “net saving” (of which the origin is most likely shorthand for “saving net of investment”). However, in our view, this terminology is not the most appropriate and could result in significant confusion for readers given that the term “net saving” in the OECD’s System of National Accounts refers to something entirely different from “S-I”. The correct terminology for (S-I) is in fact “net lending” (or “net borrowing”). Blogger Neil Wilson has been explicit on this point in the context of the aforementioned debate. In our view, “net lending/net borrowing” of the private domestic economy could also be appropriately described as the “sector surplus/deficit” of the domestic private economy.

Therefore, on the basis of the Statistics Canada glossary (which is in line with OECD terminology), the following terminology will be used in this post when referring to saving and investment:
  •  “S-I” is “net borrowing/net lending” of the domestic private sector (in the three sector model), where “S” is “gross saving” of the domestic private sector and “I” is “gross investment” (or gross fixed capital formation plus investment in inventories) of the domestic private sector. Gross saving does not include a deduction for capital consumption allowance (capital consumption allowance is known at the micro level as “depreciation”). Similarly, gross investment does not include a deduction for capital consumption allowance.
  •  “Net saving” of the domestic private sector, unlike “gross saving”, includes a deduction for capital consumption allowance. “Net saving” for the domestic corporate sector (a subset of the domestic private sector) is approximately equivalent to undistributed corporate earnings.
From the above definition, it is quite clear that households/corporations deploy their gross saving, not their net saving. In fact, it could be said that if gross saving is above zero, then it will be deployed in assets (either actively, as in the case of deployments in equipment, or passively, as when leaving it in a checking account) or deployed to reduce liabilities (e.g., repaying loans). If gross saving is above zero, deployment will occur even if net saving is zero or negative due to deduction of capital consumption allowance (i.e., depreciation). Statistics Canada drives this point home when it describes how gross saving is derived from net saving:
Added to this item [net saving] are capital consumption allowances (CCA). The latter are a cost reflecting the reduction in the value of fixed assets used up in production during the period (i.e., depreciation). Even so, they constitute available resources, since in practice, CCA is merely an accounting entry.
This is consistent with the definitions used by the OECD, and fully consistent with the approach used at the micro (business) level. Indeed, a business could in fact generate significant amount of gross saving and deploy them in various types of assets (real or financial) or to reduce liabilities while simultaneously running down its accumulated net savings (i.e., running down its accumulated retained earnings). This is why we argue that JKH’s flow equation S=I+(S-I) is really about gross saving and gross investment, as he himself mentioned when he described this equation as one of “portfolio balance” in which “the two major categories for the application of saving are investment and net financial assets”.

Gross Saving or Net Saving?

On the basis of the above, we would contend that JKH and other MMRers have themselves entertained some confusion over their interpretation of “saving” in two important ways.

First, JKH rarely specifies whether he talks about gross saving or net saving. Second, and more importantly, JKH at times seems to switch from gross saving to net saving in the same paragraph or in the same text. Nowhere is this confusion more apparent as in the comments by JKH that were published on the CNCB website. In those comments, JKH writes that:
Saving is described in proper accounting terms as funds sourced from income by virtue of being saved from income. The eventual deployment of that source of funds is described properly as a use of funds — whether such deployment and use occurs in the form of a bank deposit, a bond, a stock, newly produced residential real estate, or newly produced plant and equipment. The deployment or use of funds is separate from the act of saving itself.
In the paragraph above, JKH is referring to “deployment” and “use of funds” and, as such, ought to be talking about gross savings. Then, in the subsequent paragraph, he states the following:
In summary, the consolidated private sector account obscures, not only the view of saving as it materializes within a given accounting period in bifurcated fashion across household and corporate sectors separately, but also the view of total private sector saving as it is projected fully into the household balance sheet, when captured as a cumulative measure over a sequence of such accounting periods.
In the paragraph above, JKH should really be talking about net saving given that, after all, it is net savings that could be seen as “being projected fully into the household balance sheet, when captured as a cumulative measure over a sequence of such accounting periods”.  On this point, we think that JKH has unfortunately erred in his flow-stock reconciliation.  Flow is about the deployment (i.e., use of funds) of gross saving, as mentioned above. Stock is about cumulative net saving, as capital consumption allowance (depreciation) is netted out from both the asset side and the equity side when reconciling the flow to a balance sheet.  As you can see from the chart below showing corporate gross saving, net saving and net lending, from an empirical standpoint, this is not a minor point. Indeed, using Canada as an example, there is a very significant difference between the level of corporate gross saving and net saving in the national accounts (click on chart to expand).

Source: Statistics Canada and authors' calculations

Granted, the difference between gross saving and net saving will vary greatly between industries. For instance, in the service industry such as banking, the difference could be relatively small, but in capital intensive industries such as car manufacturing or oil and gas extraction, the difference could be very substantial.

Similarly, JKH repeats this error in his paper by again not specifying whether he refers to net saving or gross saving and by confusing the act of deployment of gross saving with net saving:
In summary, saving is a subset of income. It is a flow, not a stock. It is the residual of after-tax spending on consumption. (In the case of corporations, it is undistributed profit after the payment of all expenses including taxes and depreciation.) Saving is not the actual deployment of funds into asset acquisitions or liability reductions. Those events are defined subsequent to the fact of saving.
In the above paragraph, JKH is referring to the deployment of gross saving (i.e. use of funds) while also referring to “net saving” when he writes that, for a corporation, saving “is undistributed profit after the payment of all expenses including taxes and depreciation”.

Concluding remark 

Is JKH aware of the conceptual shortcuts he has employed when talking about the term “saving”? He is. At least one of the authors of this post has had exchanges with him on blogs about this issue and he has always been gracious and meticulous in his responses. So we are not about to claim that JKH does not know how to do proper stock-flow reconciliation.

Did JKH simplify his savings terminology so that his message about saving versus net lending is easier to assimilate for a general audience? We would say yes. And herein lies the paradox of academic expression. As a general case, we should always strive to be as precise as possible in terms of terminology or expression. But the necessity to get your message across to a general audience is bound to collide with our utopian goal to be perfectly and academically precise in the use of words.

So, in conclusion, we would argue that in trying to convince us that “saving” is different from net lending for the private domestic sector, JKH ended up using arguments that could be viewed as counter-productive in that they blurred the distinction between gross saving as it is deployed (use of funds) and recorded on a cash flow statement, and net saving as it is accumulated (as a stock) and recorded on a balance sheet.


  1. While I appreciate your figuring out the terminologies, .... in any definition of saving (Gross Saving or Saving Net of Consumption of Fixed Capital), it is still untrue that it is impossible for the private sector to save if the government runs a balanced budget! (Closed economy for simplicity).

    For the private sector can have BOTH Gross Saving as defined by SNA or the Federal Reserve FoF and Saving Net of Consumption of Fixed Capital to be positive with the government budget in balance or surplus. So not "impossible" per se.

    I went through this here


    So in the years 1998-2000, the US private sector had both positive gross saving and positive saving net of consumption of fixed capital, in spite of the budget being in surplus and the current balance of payments in deficit!

    1. Perhaps it would be helpful to also state what it is that the private sector cannot do when the govt runs a balanced or surplus budget (closed economy for simplicity)?
      You know, to put things in a larger context to avoid misuderstandings or misuses like undermining basic MMT tenents.

      Perhaps an expansion of this -

      "Of course, this meant that the private sector financial balance is negative and this you can see in line 43 highlighted in Red."

      - might provide helpful context to the more casual reader.

    2. Well yeah. And I have a blog in which I regularly argue that it is ideology to balance the budget. My post referred about was to pinpoint the issue and not digress.

      Maybe could have done differently but I like short posts. But will keep this in mind.

  2. This is excellent, thanks.

    We did have a discussion about this, and I think the thing for me to do is to take it more into account now, and figure out the best way forward with terminology.

    I think you know this, but just to explain, I borrowed the phrase “net saving” from MMT’s use of the same, and strove to do the contrast of saving and “net saving” at that MMT terminology level, without great regard for adjacent terminology at the different level of capital cost allowance.

    And I think you understand that the primary issue for the post was that of the confusion between total saving and net saving as that terminology is used generally by MMT and as it pertains to the netting of the flow of physical investment from the flow of total saving within a given defined sector.

    But I agree that distinction needs to be captured more clearly in terminology that can translate consistently to the other level of netting you have noted, which is that of the depreciation of the existing stock of investment against the current flow of income.

    This is obviously not an insurmountable task. There is a measure of saving at the level of GDP and gross investment that needs to take into account depreciation as a source of funds, in addition to undistributed business profit and household saving. And there is another measure of saving at the level of net investment that need not take into account depreciation in order to have the same matching concept. So I believe there are two parallel measures that need to be consistent in terminology.

    However, the consistent delineation of these parallel saving/investment flows is a somewhat separable conceptual issue from that of distinguishing between saving that is matched to physical investment and saving that results in inter-sector financial claims. My purpose was to highlight the latter. In doing so, I placed secondary importance on the former, with the resulting terminology overlap you note. But just remember that the origin of the terminology as it is used to apply to the general netting of investment flows I against total saving S is MMT (certainly prior to Neil’s work), which is where I was trying to draw the priority.

    I think the truth of the matter is that I was quite aware of these distinctions, but had them only in the back of my mind as I was writing the piece, while focusing on the frontal issue of netting at the level of gross investment flow rather than depreciation of investment stock. I doubt I was terribly confused when referring to the projection of cumulative investment onto household balance sheet net worth or equity. I’ve hung around corporate income statements and balance sheets long enough to know roughly how that works. But I take your point that a flow of funds statement definitely reflects gross flows as you note. So yes, there’s definitely potential for terminological confusion in combining these two different netting ideas.

    “Did JKH simplify his savings terminology so that his message about saving versus net lending is easier to assimilate for a general audience? We would say yes.”

    Thanks. I think what I did was probably more of a prioritization of a particular level at which the concept of “net” was used, without being duly sensitive to the potential for overlapping uses of that same term in a different and important way.

    This has been a pretty quick response on my part, and I know you’ve been very careful with your post, so I hope I haven’t made further confusion in the process of reviewing the language. I do take your point. So I’ll work on it. Thanks. It’s an excellent point and a fundamental one in terms of the desirability of consistent terminology across all of this netting language going forward.

  3. Additional point:

    In fact, (S – I) can be calculated at the level of either gross or net saving/investment (national accounts), and the result will be the same. So in the sense of that end result alone, there is no confusion in the netting of I from S. It can be a netting of gross flows, or a netting of net flows. Provided one is aware of the terminology overlap, the chosen level doesn’t much matter to the primary point of the difference between saving and saving net of investment (as opposed to the difference between saving and saving net of investment stock depreciation).

    Also, flow of funds generally tracks gross flows. But the balance sheets that flow of funds connects net out the effect of depreciation on both the investment stock and equity accounts. So a consistent concept of net flow of funds exists, should one want to define it that way.

    So there is consistency in all of this at the conceptual level, despite the overlap in “net” terminology.

    But I still take your point. And additional clarity is probably more important when one works corporate profits into the conversation, as you alluded to in the final quote provided.

  4. Joseph Laliberté27 March 2012 at 08:16

    Thanks for your reply. I will get back to you with a more detail reply later, but to respond to your latest point, to my knowledge there is no such thing as "net investment" in the national accounts (at least in the case of Canada). So talking about "S-I" as meaning "net saving minus net investment" adds fog to an already foggy debate as it assigns an entire new definition to "I" which is outside the scope of the official definition. Futhermore, I think you made mention in the past of the fact that "saving funds investment" or something to this effect, would you really be at ease saying that "net saving" funds "net investment"? I think you would agree with me that the proper expression is gross saving funding gross investment.

  5. Joseph Laliberté27 March 2012 at 16:37

    I will let circuit comment more on this aspect, but I think the “net saving” terminology to describe “S-I” originally came from Godley, and that MMT adopted it straight from him. In fact, it is my understanding that Godley designed his own terminology/approach to solve the issue of inconsistent terminology between countries. I would note that Mario Seccareccia and other Canadian post-Keynesians use “net borrowing/net lending” to describe “S-I”. In any case, this is a minor point in the context of the more critical issue pertaining to the confusion between the “source” and “use” of funds (corporate gross savings that are deployed) and undistributed corporate earnings (corporate net saving, as per StatCan/OECD terminology).

    On the latter point, I would say that there is (1) the issue of terminology confusion and (2) the issue of how one interprets the economic relationships (or dynamics) on the basis of this very terminology.

    I will leave aside if you don’t mind your notion that “a consistent concept of net flow of funds exists, should one want to define it that way” as this is not an exercise (I hope!) in redefining what a conventional cash flow statement is.

    I have seen comments from you and maybe other MMRers where you mention that the private domestic sector is really about optimising or maximising S (i.e. satiating its gross saving desire). On this, you have been highly critical of MMT as it sees the private domestic sector as driven by its desire to accumulate net financial assets (accumulate “S-I”). You also suggested that this kind of presentation may fit MMTers’ normative view (or something to this effect; maybe you were referring to their view on the role of government (?)).

    Now, let’s take the corporate sector seeing that you are quite familiar with corporate accounting: would you really argue that the corporate sector seeks to maximize the present value of corporate gross saving (plus dividends, since these are netted out when calculating gross saving)? OR, would you say that it seeks to maximize the present value of its free cash flow, which can be approximately defined as “Corporate Gross Saving minus Corporate Gross Investment” (plus dividend)? I would respond the latter. Arguably, from this perspective, MMT’s position on this is still on solid footing.

    1. JL: Not sure about the origin of the expression, but your take sounds about right. I should say that, although I've been influenced by the work of Godley, I've never been overly fond of the expression net saving (or saving net of investment) instead of the recognized net lending. In any event, although I would really appreciate if everyone would stick to standard terminology, this point is really secondary to the discussion on the distinction of net and gross saving (as per OECD terminology) and its implication for macro analysis.

    2. Joseph,

      Don't think any such claim about maximising is made. If so, I'd like to know.

    3. Joseph Laliberté27 March 2012 at 20:47

      Don't have time to browse through all the comments, below are two examples:

      From JKH, response to Steve Waldman. (republished in its paper)
      "To the degree that MMT associates saving with (S – I), it leverages the message of its normative view, that the private sector is driven by the motivation to satiate its desire for net financial asset accumulation. This general view has great associated consequences for its normative view on deficit financing, etc. etc.
      This can be problematic from a logical perspective, and therefore from the perspective of normative balance.
      It is not clear at all that the desire for net financial assets is in fact the driving force, in my view.
      It is more the case I think that the true force in this Keynesian context is the desire for saving per se – i.e. for S.
      And S can be mapped into two components, as above, “I” and (S – I)."

      JKH again, in comments at the MMR blog on the post about the Krugman graph:
      "And CP then says:
      “If you want to maximize the strength of the private sector, you maximize S. Maximizing S is not the same as maximizing S-I.”
      That’s right as a statement about maximizing net worth, and very insightful."

    4. Joseph,

      I believe the first quote has no maximizing.

      The second one you quote refers to a quote by "CP" and is about policy per se, not the individual sector's behaviour about maximizing itself. It is about the government attempting to maximize the saving of the private sector given constraints and not maximizing "S-I" in any definition of S/I.

  6. Joseph,

    Net Investment is standard terminology. See NIPA's article below on government's finances:

    Table 3, Footnote 4


    1. Ram, it's true the concept is a recognized standard. For instance, the OECD defines it as net fixed capital formation. Although, that's really an irrelevant point, if you ask me, for reasons correctly pointed out by JL in his first comment. JL is also correct that the Canadian statistical authority doesn't define it for its own national accounts purposes.

    2. Nonetheless Gross Fixed Capital Formation is used by Statistics Canada to indicate consumption of fixed capital is not subtracted.

      This automatically implies a definition Net Fixed Capital Formation whether it is used by Statistics Canada or not. But you did say OECD's handbook uses it.

      Whether the Canadian statistical authority defines "net investment" is a bit irrelevant ;-) This is because the SNA is very focused on production and the term investment is less appropriate.

      We are not going into such levels of detail, so the difference between capital formation and investment is not appropriate.

      So there is no confusion created by the usage of net investment.

      In fact at a sectoral level, net saving minus net investment is a reasonable definition. (these nets are net of CFC :-)

      While the underlying theme of this post is that things can get technical,

      More generally, the way national accountants do it is the net saving and consumption of fixed capital are both sources of funds. That's different from saying "only gross saving is a source of funds and not net saving" if that's the claim.

      Before the closing stocks are calculated, consumption of fixed capital makes an appearance again via revaluation of assets and liabilities.

      So, gross saving is indeed added to net worth as can be seen from UK Blue Book, but unlike claimed by Joseph in MMR blog. It then gets subtracted at revaluations.

      But for calculating profits consumption of fixed capital is subtracted.

      So JKH had been using saving correctly with saving net of consumption of fixed capital in mind. The fact that JKH didn't specify he is talking of this instead of the gross measure is a minor point really compared to the huge debates we have been having.

      To summarize, the focus of SNA is on production. However, in order to discuss flow of funds, something more than production itself needs to be used and I see net saving minus net investment as the more appropriate measure.

      On other points, JKH said that net saving (i.e., net of CFC) is a source of funds, but he didn't say CFC is not a source of funds!

    3. Joseph Laliberté27 March 2012 at 21:23

      I have seen you many times making the point that JKH has been implicitly clear all along that he was talking about "saving net of consumption of fixed capital". But just as a reminder, JKH provided the following comment on the Krugman's graph (btw- the Krugman's graph is about gross saving and gross investment):
      "S = I + (S – I)
      The blue line = S
      The red line = I
      The gap between the blue line and the red line = (S – I)
      Oh, yeah … and I’m pretty sure he knows what the definition of saving is.
      Anybody ever see a neat graph like this in MMT land?
      I.e. one that actually includes I?”

    4. Maybe that's what I thought initially.

      And maybe it should be made clear if it's gross or net.

      But my point was that JKH went into a definition of undistributed profits which is right. JKH said it is a source of funds, which is right but didn't mention consumption of fixed capital itself is a source of funds (rather than saying it's not).

      So if the consumption of fixed capital is important for the discussion, I am sure JKH knows how to handle this.

      However in the big post, there was no need to bring this in because the simple mental MMT model being addressed doesn't discuss depreciation itself.

      If the simple model itself doesn't talk of depreciation it's okay to just use saving instead of distinguishing gross or net because these are same.

      Anyway I shouldn't talk for JKH!

    5. If your criticism are more like a referee's criticism on a paper rather than another paper criticizing the paper, then I am okay with you.

    6. As far as commentaries go on this issue, the paper contains the relevant background information. It's also worth reading if you're interested in knowing JKH's views on this topic, as we mention above.

  7. JKH, thanks for your comments. I apologize for the delay. On my part, I wish to add that our aim here is really to try and clarify statements that we felt might be inconsistent with our understanding of the concepts of (net) saving and gross saving (and its use and source). We believe the motivation that led JL and I to write this post is one that you may actually relate to. In fact, while writing this post, I was reminded of an exchange you had with Scott Fullwiler several months ago, in which you expressed reservations about the manner in which MMT was being presented. JL and I also shared this view, which then led to us to examine some commonly made statements (and the use of certain terminology) and try, as much as possible, to compare it with conventional or established terms of art.

    Again, this post is not meant to challenge what you wrote in your paper (as we mentioned above, we actually recommend it). Rather, we are trying to clarify these issues in the spirit of hopefully settling on a common understanding of these aspects moving forward. Thanks again for your valued comments. We recognize that you place great emphasize on precision and clarity in your responses.

  8. Ciruit @Mar 27, 2012 03:11 PM,

    Btw the terminology NAFA and net saving comes from the New Cambridge group of the 70s - Godley/Kaldor/Cripps.

    More likely they were probably was old British national accounting terms (NAFA old is different from the NAFA used by accountants these days!)

    "The PSBR in any year can be defined as the public sector’s net de-cumulation of financial assets (net dissaving) which by accounting identity must be equal to the net acquisition of financial assets (net saving) of the private sector, home and overseas; ...."

    from the Scourge of Monetarism by Nicholas Kaldor.

    Quoted by me here as well.


    Also I'd like to know which Post-Keynesians use Net Lending/Net Borrowing in sectoral balances - if you have a reference. Just interested.

  9. Just to clarify, I do not have a problem with using the terms "saving" or "net saving" (in the MMT sense) if there is a reference somewhere to "S-I", net lending, saving net of investment or any other format (whether it be in text or chart form). Ram, you may wish to check out Mario Seccareccia's work.

  10. This all seems like a pretty big storm in a tea cup. Saving is a flow, equal (essentially) to disposable income not spent. In a closed economy, with government sector in balance, this happens to equal investment, I. Now, one could adjust this saving/investment to account for capital depreciation, in which case one could introduce the distinction between gross and net saving, but this is no big deal, conceptually. Similar to the distinction between GDP and NDP.

    What is misleading though, is the occasional tendency of MMTers to use the term "saving" to describe the net accumulation of financial assets by the private sector (S - I ). This is unhelpful because we are really talking about a stock adjustment, and "saving" is best used to describe flows. I concur that calling ( S - I) "net borrowing", or "net lending", is much clearer, as it implies the notion of a stock adjustment.

    I find it unfortunate that some use the term "saving" or "net saving" when what they really mean is "net loss/accumulation" of financial assets by the private sector. Even worse when it is argued that "saving" can only occur when the govt runs a deficit. This is rhetorical mischief, or just sloppiness.

    The private sector (firms) can invest and households save and accumluate real assets without any need for a budget deficit, or trade surplus. The stock of REAL assets rises. But in terms of FINANCIAL assets, it is a wash, as firm liabilities offset household ownership of equity/bonds/deposits (depending on how firms finance themselves).

    1. rjw, welcome and I appreciate you checking out the site. Your use of the expression 'rhetorical mischief' is interesting. JL and I actually discussed whether there is anything to this. But as you allude to in your commentary, it would be hard to say for sure.

    2. it is argued that "saving" can only occur when the govt runs a deficit. This is rhetorical mischief, or just sloppiness.
      Financial Claims Made Simple

    3. Yup - I tend to stick to the national accounting terminology. Net saving is not the same as saving (unspent income).

  11. Vassilis Serafimakis2 July 2012 at 12:11

    It's MMTers and MMRists. Let's be clear on the terminology...

  12. Hi,

    Did you write an article about savings=investment, and how a rise in household savings may just raise inventories? I did some searches, but I could not find the article.


  13. This, by the way, is a highly infectious blog.... I mean disease to mammals. Please Visit:

    accountants in barking
    payroll services in London