Imagine that over the next week (in a closed American economy — the rest of the world has never existed) everyone sold all their financial assets, paid off all their debts, and deposited the remaining money (and any currency they have) in their checking accounts. No money-market funds, even. Just banks with reserve accounts at the Fed, holding everybody’s money in “cash.”
All those other financial asset prices would dive to zero. Late sellers would sell for nothing.
Would the remaining money in all the bank accounts equal U.S. government debt? That seems to be the implication of MMT thinking, because the remaining money only exists because it got spent into existence by the government deficit spending — crediting bank accounts with that fiat, ex nihilo money in the first place.
Net financial assets = gross financial assets = government debt
(If the government had always just deficit-spent instead of borrowing to cover its deficits, “government debt” would be replaced by “cumulative to-date government deficit spending.”)
I ask not just for clarity, but because (as always), I’m struggling with the relationship between fixed assets and financial assets, between saving and investment.
It’s said that the true wealth of the nation — the “national savings” — consists of its real assets: stuff that can be consumed in the future through use and time/natural decay. The NIPAs only count “fixed assets” — hardware (equipment), software, and structures, so let’s pretend that those constitute all real assets (which they don’t in actuality — not by a long shot). Net investment — purchases/creation minus consumption of fixed assets — increases the stock of fixed assets/”savings.”
In theory, financial assets are just financialized, monetized representatives, proxies, for the real, fixed assets that underly them. And indeed over the (very?) long term, the quantity of fixed assets and net financial assets rise together. Both are much larger today in the U.S. than they are in Thailand, or the U.S. in 1910. Financial-asset values wander all over — even over decades — based on “animal spirits,” but again in the long term…
If that’s so, then in our thought experiment:
Net financial assets = gross financial assets = government debt = fixed assets
The quantity of fixed assets increases over time through net investment. But by MMT thinking, net financial assets can only increase through government deficit spending (or trade surpluses). What is the mechanism whereby government deficit spending is translated into more net financial assets that embody the increased stock of fixed assets?
I imagine a necessarily political mechanism something like the following:
1. People and businesses buy/create fixed assets, resulting in more economic activity — creating/consuming, buying/selling, spending/income.
2. Those increased quantities (both stocks and flows) create more demand for government services. Both individuals and businesses would be decidedly unhappy, I’m thinking, if today’s government were the same size it was, at least in absolute terms, in 1870. (Conservatives and libertarians may say otherwise, but they’re talking through their hats.)
3. Legislators and executives who don’t provide those increased services don’t get re-elected.
4. Taxation lags behind spending — resulting in deficits — because A) people hate taxes and vote against politicians who raise them, and B) if deficit spending is not sufficient to match the increases in fixed assets, depressions result, and the “fiscally responsible” leaders get voted out.
5. The new money from government deficit spending is used to purchase financial assets, driving their prices up to (roughly) match the value of fixed assets.
This is thinking of government and the Fed as one consolidated entity. If you think of them as separate, you can imagine a different mechanism, in which the Fed and the congress/president are engaged in a constant chicken game over inflation, unemployment, and GDP growth, to determine how and when to increase the amount of money/net financial assets (ultimately through deficit spending) to match the stock of fixed assets.
These mechanics would also explain how buying/creating a bunch of drill presses will — through a long, tangled, and messy political process, and in the long but not the short run — result in more “loanable funds.”
Cr0ss-posted at Angry Bear.
Comments
63 responses to “An MMT Thought Experiment: The Arithmetic and Political Mechanics of Net Financial Assets”
Steve — I think you’re making things a bit too hard.
First, the right way to get rid of private financial assets in a thought experiment isn’t to imagine everyone sells them for cash. To whom would everyone sell? You have to net them out. I hold your mortgage, so you give me cash, perhaps selling a fraction of your home to cover the redemption. I hold stock in Apple, Apple redeems it by handing me one billionth of its plant/software/knowhow-and-culture/etc. (To make this thought experiment work, we imagine that all assets can wrapped up into ownable, infinitely divisble bundles, which is terribly counterfactual, which is why it’s a thought experiment.)
Now all private claims have been redeemed. No one owes anything to anyone — almost. I hold bits of house and bits of the planned iphone7 in my basement, as do you. We hold different bits of things in different amounts, because we did not hold the same set of (net) claims before the great redemption. We both also hold claims against the government which, by assumption, have not been redeemed. I hold some cash and government debt. You do too.
Let’s get to that “almost”. There are still obligations in this world, because the government still exists. Let’s put aside formal claims for the moment. We’ll come back to them. Imagine the government has not issued any money or debt. Then we all secretly hold an asset — a flow of government service. We have valuable access over time to the use of roads, a fire department, an army, a dispute adjucator. We also secretly hold a debt, a tax liability. If we imagine, again counterfactually, that all assets are storable and divisible, we could extend our thought experiment. We surrender some of our bits of house and proto-iPhone7s and receive some canned interstate-highway-and-dispute-adjudication goodness. We might surrender stuff and receive services in different quantities depending on who we are and what we have and who our lobbyists are. But a distribution is determined and redemptions are arranged. We are all (well, at least both of us) even steven. We hold all the assets we will ever have, and owe nothing to anyone. Maybe some of us are outrageously rich and others will not be able to survive on our alottments. Them’s the breaks!
Now suppose some people gang up and obtain the power to coerce confiscations of real assets. Perhaps its poor people unhappy with their alottments. Perhaps its rich people who already held the bulk of canned police services and so find coercion convenient. Whatever. These people offer a product. It’s the same product that the Onceler offered in the Lorax, a thneed. A thneed is a thing that everyone needs, because it can become whatever it is you need at will. A thneed, it turns out, costs nothing to produce. It is just a promise. The people with the power to coerce give you a piece of paper that says “I’ll coerce one unit of transfers for you at will, just tell me what and when you want it in a quantity of 1 unit.”
What does a unit mean? It means whatever level of transfers you expect them to coerce as one unit. If they issue 100,000 of these units, and if you think they are willing and able to coerce transfers of 10 homes, then a unit is worth no less than 10/100,000 of a home to you. It is actually worth more than that to you, because you can control the timing and the nature of the flow. Maybe you will want more proto-iphone7 rather than some extra home a few weeks from now. It’s much better just to hold the coercion coupon. So you are willing to surrender 11/10,000 of a home for a coercion coupon. The coercers take the homey goodness, and deliver it to whoever they wish. They sell a lot of coupons this way. A small fraction of them get redeemed, and so the coercers choose people whom they dislike and from whom they can easily take and make good on the coercion when asked. Because they credibly do so (and don’t focus their coercion only on coupon holders), coupons remain valuable. In fact, once the game gets started, they may need to do very little actual coercing, because when people come to redeem their coupons, others might be purchasing coupons, so rather than going to the trouble of coercing, they can just transfer the proceeds from the new coupon issues. As long as the option value of coupons means that purchases are greater than redemptions, the coercers need coerce very little. But, there will be times when that is not the case, when more people wish to redeem then want a nonspecific option, so coercers must stand willing to coerce. Worse, if people come to doubt coercers ability or willingness to coerce, this will cause a run from coupons, forcing actual coercion big time. So coercers must regulate their issue of coupon based on their ability and willingness to coerce transfers.
You asked why government debt (coercion coupons) and the real economy tend to grow together. It is precisely this: the amount of coercers can credibly promise is related to the size of the economy. It’s easy to force a person to give up her third ice cream cone, but much harder to force coercion of her first slice of bread. In an economy where people’s endowments typically amount to a full meal and lots of ice cream, coercers can issue a whole lot of coupons. In an economy where everyone only has half a meal, it is too hard to do a lot of coercion, the cost of instilling a lot of value to “a unit” becomes higher than the benefits they get from distributing real stuff to favored parties in exchange for coupons. So coercers regulate their issue of coupons based on the size of the real economy and their perceived cost of coercion. The relationship between coercion coupons and the real economy is very loose: coercion coupons, unlike Apple stock, aren’t a claim to any particular thing. Coercers decide only when it becomes necessary to coerce what the coupons will be a claim upon. Government debt does not represent or proxy for any sort of real investment. It is valuable to the broad public, because individuals prefer thneeds to claims on specific assets that may not match their evolving wants. It is valuable to the coercers, because they earn a “seigniorage rent” from the option-value associated with their coupons and a perpetual float associated with the public’s willingness to hold some stock of coupons in perpetuity, which allows them to obtain more resources for favored purposes through this scheme than they would have obtained through direct coercion to their friends. Plus, since a service (thneed provision) is attached to the coercion, coercion is much easier and therefore less costly to perform than would be naked confiscation on behalf of friends.
Does this make any sense? Do you think it’s right?
Steve,
This is an interesting analytical challenge you’ve presented.
Warning: the solution I offer here reflects my strong belief that the economics profession is beholden to the logic of accounting more than it appreciates at this stage. MMT understands this. In my opinion, it’s the seminal contribution of MMT (and its originators and forbearers) to economic analysis. That said, I am not an MMTer.
Given – you’ve assumed a closed economy with banks, and no other financial assets.
The first objective is to identify the destination. The general nature of the journey to the destination can be worked back by inference.
AT THE MARGIN, the conditions you’ve prescribed implicitly include the “no bonds†version of deficit spending; i.e. the government spends; bank deposit liability accounts are credited, and bank reserve accounts are credited. The government issues no debt. Cumulative government deficit spending is forced through the banking system.
Assume the cumulative government deficit is $ 10 trillion.
(Note: this is not far from the actual US case. The nominal outstanding debt is closer to $ 15 trillion. But that includes internal accounts within the consolidated government, such as social security, other trust funds, and debt held by the Fed. That said, the share of debt held by the Fed still represent deficit financing, since it is offset by Fed liabilities held by the public. Anyway, I’ll assume $ 10 trillion as a round number is closer to the cumulative deficit than $ 15 trillion.)
Suppose the scenario (i.e. “destinationâ€) represented in your post is achieved by a series of transactions starting from status quo. I’ll refer to the process as “Financial Conversion†or “CONVERSION†for short.
Conversion means you’ll end up, for one thing – at the margin, initially – with an additional $ 10 trillion of deposits in the banking system, offset by $ 10 trillion in bank reserves with the Fed. That’s the result of Treasury redeeming all bonds for cash upon Conversion.
The facilitating Fed/Treasury accounting arrangement for conversion is a detail – just assume the $ 10 trillion is transferred internally from the Fed to Treasury. Alternatively, assume that Treasury and the Fed are combined into a single government agency that takes on the operational currency issuer function.
That $ 10 trillion is the NFA position (technically, unchanged from what it was). But it’s got nothing to do with real assets.
(Also, for simplification, assume away any real assets held by the government. You have to simplify in order to get to the root of a core answer here.)
So we assume real assets are held entirely by the private sector in this closed economy.
Let’s assume real assets total $ 40 trillion in value.
Assume a split of $ 20 trillion held by households and $ 20 trillion held by business.
Every household and business unit has a balance sheet. Post-Conversion balance sheets include only two types of assets – bank deposits and real assets. They include only one type of balance sheet entry on the right hand side – net wealth or equity. There is no debt on either the asset or liability side of balance sheets.
Conversion includes some kind of iterative trading process whereby all units make the transition from their old balance sheets to their new balance sheets.
So, as suggested, suppose the very first step in the Conversion is the government’s transaction – i.e. the first step is that the government swaps $ 10 trillion of existing publically held debt for $ 10 trillion in bank deposits. This provides $ 10 trillion of additional bank deposits AT THE VERY START to facilitate the full asset trading process that is subsequently required.
That asset trading process involves both financial assets (with mirror image liabilities) and real assets.
Now – to the meat of the solution
But here is a preliminary required detail that follows from my opening statement:
……………….
I think that you’ve made one and possibly only one critical error in your analysis to date as I’ve seen it. And it has to do with accounting logic.
The best way to illustrate this is a simple example of the point in question.
Suppose a household owns a common stock. Suppose that stock is the only asset it owns. That stock is classified in general terms as an EQUITY CLAIM. It is a financial asset.
That household has a balance sheet in financial terms. There are only two entries. On the left hand side is the EQUITY CLAIM. On the right hand side is EQUITY. Those are two different “thingsâ€. The LHS is a financial claim. The RHS isn’t. It’s a measure. It’s a measure of net wealth.
In the taxonomic paradigm of “net financial assetsâ€, the LHS is a gross financial asset, and the RHS that is a measure of NFA. This is a very important logical point.
It is ALSO the case that the RHS represents cumulative saving, or savings. In this example, the LHS is a financial asset and the RHS is a net financial asset position. NFA is one variation on cumulative saving or savings.
Anyway, the main point here is to distinguish between EQUITY CLAIMS and EQUITY, the latter as a RHS balance sheet entry.
From there, one can see that businesses both hold and issue equity claims. The NFA position of a business balance sheet that holds a stock and issues an equity claim against it is zero.
But the NFA position of a household that holds a stock and has RHS equity is a positive NFA claim equal to the size of either side of its balance sheet.
One can then extend the equity cases to those of debt on either side of the balance sheet and group both equity and debt claims (LHS positive and RHS negative) together in summing to an NFA position that will equal RHS equity.
And from there, one can add real assets that have no effect on NFA, but that do affect RHS equity.
That’s the guts of balance sheet accounting.
I hope you’re getting the drift of what I’m going on about here.
Finally, a government that issues debt of D (externally, consolidated) has issue NFA in the amount D. Assuming away the value of any real assets it holds (for simplicity), that government has implicit equity position of (D). That equity may be viewed either as a negative RHS entry or a positive LHS entry. (If Steve Waldman is reading this, it’s a point I’ve been revisiting and going on about at his blog for several years now.) These equity interpretations for government are essential if one is to make full and cohesive sense of the meaning of national saving, among other things, as noted further below.
So those are the balance sheet accounting preliminaries that are required for the rest of my explanation.
………………………………..
Getting back to the Conversion trading process, some examples of transactions required for the economy to reach the destination of its new balance sheets through Conversion:
– The first step already mentioned, which is that households (and businesses) with old government bonds exchange them for money and bank deposits (remember that the closed economy assumption means that bonds that otherwise might have been previously held by foreigners were instead held by households and businesses and domestic financial institutions)
– In our example, that first step injects an extra $ 10 trillion of bank deposits into the financial system, liquidity that will help with subsequent Conversion requirements
– Households trade other financial assets for money that ends up in bank deposit accounts; counterparties are obligated under Conversion to redeem those financial assets with money
– Households can pay off debts partially or wholly with money from those sources
– Households with no debt remaining are set with their new asset mix of bank deposits and real assets (mostly real estate) and 100 per cent RHS equity
– Some households may have to sell real assets to end up with positive RHS equity; e.g. if they have a large mortgage
– Some may end up with no real assets and with a balance sheet that consists of bank deposits as assets and RHS equity
– A potential problem is households that have negative RHS equity to start – i.e. net debt; that gets resolved through bankruptcy
– Businesses go through a similar process, having to balance their eventual asset positions of money and real assets against RHS equity
– And non-bank financial institutions go through a similar process, where they are left only with residual real assets and bank deposits and RHS equity; they are then liquidated since as non-banks in the new regime they have no function
– You assumed no central bank notes are outstanding by your specs
– Banks must also adjust their balance sheets; their loans are paid down; they are left only with reserves ($ 10 trillion), real assets, deposit liabilities ($ 10 trillion), and RHS equity (initially equal to real assets).
Anyway, the end result of all of this is the following:
The economy will have total assets of $ 50 trillion – $ 10 trillion in NFA and $ 40 trillion in real assets. This is offset by $ 50 trillion in RHS equity.
(One issue is whether equity claims (i.e. stock) would continue to exist, notwithstanding the unwinding of all debt. This could make big difference to the liquidity of trading operations required to complete the conversion trading process. If so, we can add equity claims to the list of eligible assets for both households and businesses, and the same amount of equity claims to the RHS side of business balance sheets. That makes resolution of final balance sheets in total much more liquid as a process. Without them, bid offer spreads on trades in both real and financial assets would be much wider than with them. Assuming equity claims are allowed, those claims will exist on the left hand side and/or the right hand side of various balance sheets, such that they will net to zero in aggregate. Equity claims are gross financial assets, but not net financial assets in aggregate. Finally, it would seem that equity claims are essential in the case of business, and for continuation of a capitalist economy, and therefore those claims must be held as financial assets by somebody – e.g. the equity claims of banks as noted above.)
In conclusion, I believe the economics profession in general will eventually have to bring itself more up to speed on the importance of fully coherent financial accounting that aligns properly with theory that can connect to the real world. Financial accounting requires an understanding of three different types of financial statements – income, balance sheet, and flow of funds. These apply at both micro and macro level. Macro economists for the most part seem to be expert in the macro income statement – NIPA in this case – somewhat less so in reading balance sheets, and relatively unaware of the importance of the entirely separate flow of funds statement. For example, somewhat endless discussions about revising the definition of “saving†can be attributed directly to a lack of appreciation for the essential purpose of a separate flow of funds statement (often called sources and uses of funds at the micro level). Economics education programs should require rigorous training in the basics of financial accounting, in my view.
Regarding your basic point about identifying what saving is in the context of all this:
In the above example, in a closed economy, the cumulative saving of the private sector is $ 50 trillion, marked to market (assuming the value of real assets is marked to market). (This assumes for simplicity that the conversion process hasn’t destroyed value.) NFA is $ 10 trillion. Cumulative investment is $ 40 trillion (marked to market). Cumulative national saving (again marked to market) is $ 40 trillion (NFA is netted against the same amount as a “liability†of government in the form of reserves). These cumulative numbers are all versions of “savingsâ€, which is the cumulative result of saving, usually marked to market.
The corresponding “saving†numbers are those generated per accounting period. For example, suppose the government runs a $ 1 trillion deficit and the economy adds $ 2 trillion in real assets, both in a given year. Then incremental NFA is $ 1 trillion, incremental investment is $ 2 trillion, incremental private sector saving is $ 3 trillion, and incremental national accounts saving is $ 2 trillion.
Thanks Steve. Amazing. I’ve just gotten home and am going out again, will get back to it tomorrow, but on first scan:
“EQUITY CLAIMS and EQUITY”
I’m liking this a lot. That in itself is clarifying things.
More anon.
P.S.
I assumed no loss in the market value of real assets in my example above.
This contradicts the specs in your problem, which was deliberate on my part.
My representation is a base case, to demonstrate the accounting.
Your spec that people starting “selling†assets at zero value could be treated as a derivative of this base case.
The difference between original market values and CONVERSION market values could be viewed as the cost of trading in an illiquid market.
That said, the economy receives an injection of $ 10 trillion in bank deposits at the outset in my example. This circumvents the need otherwise to repo Treasuries to get liquidity, since that would not be allowed in the new credit free system. This $ 10 trillion therefore adds considerable liquidity for the CONVERSION process.
Nevertheless, I dealt with a similar case of real assets trading at zero value in the following comment, several days ago:
http://worthwhile.typepad.com/worthwhile_canadian_initi/2012/01/steve-landsburg-goes-meta-on-me.html?cid=6a00d83451688169e20168e54d1e2c970c#comment-6a00d83451688169e20168e54d1e2c970c
In my example for your post here, even if the market value of real assets becomes identically zero at CONVERSION, the nominal value of NFA remains $ 10 trillion. This is based on the assumption that the nominal accounting value (after CONVERSION mark to market) of real assets can’t go negative (micro or macro), and that the value of RHS equity for the private sector would become the $ 10 trillion value of NFA (after the $ 50 trillion in private sector RHS equity is written down by a $ 40 trillion loss in real asset value).
Again, the allowance or non-allowance of equity claims is an important caveat to the operation of your scenario.
The non-allowance of equity claims would have several fundamental consequences.
First, conversion trading would become more illiquid, depressing asset values. Those firms with real assets exceed the value of their RHS equity would have greater difficulty resolving their balance sheet positions in the face of having to liquidate credit. By being able to issue additional equity claims as part of that resolution trading process, CONVERSION liquidity is improved.
Second, the coerced extinction of existing equity claims is an enormous problem, likely provoking the collapse of the capitalist system. I don’t know if that’s what you had in mind here. But if firms must liquidate the equity claims they have previously issued against their RHS book equity positions, there is no “there†there anymore in terms of ultimate ownership of household capital claims on the business sector. Households at best end up taking existing business real assets piecemeal onto their own balance sheets in exchange for their old equity claims. Apart from the problem of determining granularity of real asset distribution in that process, one must then visualize how existing production processes might proceed from there, in the absence of some sort of evidence of shared financial claims of household on businesses. But that condition is precluded by the specs of your post. And that might imply that real assets end up at closer to zero value on average, with the economy shrinking its real asset based production dramatically. I haven’t absorbed Steve Waldman’s comment in its entirety yet, but maybe this is an area he has explored.
P.P.S.
SAVING is NEVER “equal to investmentâ€.
In a closed economy with a balanced budget, the MEASURED QUANTITY of saving is equal to the MEASURED QUANTITY of investment.
But that equivalence is one of measure, not of substance.
Investment can’t be negative. Saving can be, at the sectoral or agent level.
In general, savings (plural) manifests itself as RHS balance sheet equity.
It is a pure measure, with no real substance.
Real investment, a LHS balance sheet entry, is substance that is measured for balance sheet purposes.
That is true for micro and macro level balance sheets.
More of this here (which you’ve seen I think):
http://worthwhile.typepad.com/worthwhile_canadian_initi/2012/01/why-saving-should-be-banned.html?cid=6a00d83451688169e2016760653ef4970b#comment-6a00d83451688169e2016760653ef4970b
That comment was received, to put it mildly, unenthusiastically. Saving was described as a “non-thing†by our host there. If so, then income is also a “non-thingâ€, because saving is a subset of income. I prefer not to go down that rabbit hole and end it there. Saving is a pure measure. It is not “equal†to anything of substance, to be sure, and that is the intention in double entry bookkeeping. Savings (plural) is equity, which is readily accepted as a conceptual residual in accounting.
Apologies if I’m overdoing the commenting here, but you seem to be very interested in exploring the overall theme lately. Food for thought, maybe.
P.S.*3
Sorry, one more.
With reference to a popular head scratcher (apparently), the sale of an antique is NOT included in income and therefore not in saving either. It can’t be. To do so would perpetuate double counting of the same value in income and saving, over and over, depending on the velocity of transactions in the same antique over time.
(What IS included is the income generated for any broker dealer who acts as agent in the transaction. That income is an expense for the seller.)
The antique itself is an asset, swapped for cash by the buyer. That is an exchange of equivalent value, generating no income. It may result in a capital gain or loss on the seller’s books, but that is not included in income for purposes of reconciling macro GDP/GDI.
E.g.
Book value of antique $ 10,000
Selling price 20,000
Capital gain 10,000
Seller’s income otherwise 50,000
Seller’s income from auction 0
Auction Commission paid 500
Income to Auction Dealer 500
Net change to GDP/GDI 500
(There is no net change to aggregate GDP/GDI, compared to the counterfactual, if the seller is a business that charges the 500 as an expense. This is not inconsistent. The 500 in that case becomes a cost that becomes embedded in a downstream product for purposes of measuring aggregate GDP and income, and that’s not the case when the seller is a household.)
The $ 20,000 selling price WOULD be included for its effect on both the balance sheet and flow of funds statement for the seller.
His balance sheet equity would reflect the $ 10,000 capital gain on the antique sale.
His sources and uses of funds statement would reflect a $ 20,000 cash inflow, with uses being a $ 20,000 cash position increase, a $ 10,000 book value asset reduction (the antique) and a $ 10,000 gain to book equity.
The purpose of the sources and uses of funds is PRECISELY to capture the detail behind transactions that don’t affect income but that do affect the balance sheet, including the equity position if capital gains or losses are involved.
When savings (plural/cumulative) is interpreted as RHS equity, it will include these cumulative capital gain effects. It is possible to split that interpretation into the sum of cumulative saving plus the sum of those cumulative capital gain/loss effects. That typically isn’t sought overtly for accounting reconciliation purposes, but it should be borne in mind that such decomposition is required implicitly to fully reconcile the interpretation of RHS equity as savings.
So RHS equity as a representation of “savings” is really cumulative saving, plus those realized capital gain or loss effects in actual transactions affecting the balance sheet.
I’ve mentioned this realized capital gain/loss effect on the value of RHS equity several times in passing at points, in my comments above. But I haven’t stated it clearly enough as a point on its own. It’s not easy to achieve consistent clarity in trying to articulate the complete algebra of the accounting at this level of generalization.
Sorry, JKH, and Steve! I was scanning so quickly (was running late) that I ran the two comments together.
Am working on responses. This is challenging stuff for me, glad you find it of interest, much appreciate your deep thinking.
@Steve Waldman
“We are all (well, at least both of us) even steven.”
Wait so this whole hypol is dependent on the fact you’re both named Steve?
(just kidding) :o)
I’d point out a FRED chart RSJ posted last year that I thought fraught with significance, MZM money supply and publicly held debt are roughly the same size– its not a good sign when the two get out of whack.
http://windyanabasis.wordpress.com/2011/03/28/leaving-modern-money-theory-on-the-table/
Given the quality of previous comments, I’m sure that what I’m about to say is going to seem pretty feeble, but for what it’s worth, I’d think of it like this:
If there is no investment, in a closed economy, aggregate saving is zero.
@vimothy
only if you define aggregate saving as S + (T – G)
S is 0 with no investment and a balanced budget
@vimothy
Vimothy,
BTW, I can’t recall if you consider yourself MMT or non-MMT (sorry).
But either way, if you do believe that aggregate saving = S + (T –G), you may anticipate a knock on the door in the middle of the night at some point, when a man wearing dark clothes and sunglasses will inform that you are no longer “MMT readyâ€.
Notwithstanding yours would be the correct definition of national saving in a closed economy, a definition from which NFA and the rest of the sectoral stuff is derived.
Be careful with declarative statements on this subject.
Just some friendly advice.
🙂
Additional:
BTW,
Closed economy:
National saving by definition
= S + (T – G)
= private sector saving plus government surplus
= investment
And:
Closed economy:
C + I + G = C + S + T
I + G = S + T
S – I = G – T
Net private sector financial saving = government deficit
Or cumulatively,
NFA = debt
And since:
S = I + (S – I) by tautology,
Private sector saving = saving required for investment, plus net financial saving
“Saving required for investment†is NOT the same as investment
In a monetary economy, one is a monetary item; the other is a physical item
That’s the reason for “required forâ€
All the equivalences here are ones of measurement, not substance
(That’s an important distinction in the overall logic)
@JKH
JKH,
Lol, not MMT–more of a (somewhat) interested observer–but if I must I stand ready to meet them at my crossroads.
Anyway, yes, that’s the definition I use. Ignoring the open economy case for simplicity, aggregate or national saving sums public and private. A deficit matches private saving with public dissaving exactly, thus its sum is zero. (This relates of course to our recent discussion at WCI).
Not sure why this should be so controversial amongst a group that prides itself on the rigor of its sectoral analysis. For some reason, they seem allergic to the whole notion of public saving. IIRC, it doesn’t exist, which has never made much sense to me. That said, if you spend enough time in the comment sections of MMT blogs, you do come across some pretty crazy stuff…
@Steve Waldman
Hey Steve:
As so often, I feel somewhat overwhelmed by your understandings, so I don’t think I can provide a really coherent or on-point response. (I’m feeling the need for pleasant surroundings, a whiteboard, a bottle of scotch, and some lengthy discussion.) This may just be catching up to where your thinking begins. But here’s where it’s taken me.
Perhaps obvious but a clarification for me: real assets and claims on real assets are completely different things. Robinson Crusoe’s island exists; we could call it an asset. It’s *value* is determined solely by the island’s ability to satisfy Crusoe’s wants and needs. His claim on that value (his “ownership”) is determined solely by the fact that he’s alone, and has no competing claimants. When Friday arrives, both the value and the ownership of the island change, and those are determined by Crusoe and Friday’s relative power positions — ultimately, their coercive power.
So nobody actually “owns” a real asset; that’s at worst a meaningless concept, and at best shorthand for “a legal right to control that asset.” The claim could carry an infinite variety of rights (and limitations on those rights) as to what one can do with the asset, hence the human value one can derive from it.
Some of those “ownership” claims are specific and redemption is legally enforceable (the title/deed to your house gives you an enforceable claim to live in it). Others (i.e. dollar bills) are non-specific, and redemption for any other specific good is non-enforceable — you can’t force me to take your dollar bill for my apple.
(The one exception for dollar bills is taxes, though the required redemption is self-imposted: the government must redeem that dollar bill for the “good” of my not being coerced/jailed for tax evasion. The real asset being claimed is freedom. That’s a tangent I won’t follow, even though it might be necessary here.)
(I think you talked about required redemption in a post a while back; I need to get back to it.)
So:
“Now suppose some people gang up and obtain the power to coerce confiscations of *claims on* real assets.”
“A thneed is a thing that … can become whatever it is you need at will.”
So a thneed has two “values”: 1. its nonspecific notional exchange value, and 2. coercive power to force any exchange for any *specific* claim I designate. There are no limitations on the rights it imparts (rather like having a gun to every other person’s head).
“government debt (coercion coupons)”
Removing the balance sheet complications of government debt, assuming government has just issued dollars: I think you’re saying that dollars are thneedles/coercion coupons.
But on the face of it, it seems that dollars don’t deliver the second right. I can’t force you to take my dollars for your apples.
But more realistically: the extent to which dollars deliver the second right — their full thneediness — is dependent on the particular situations of exchange. If some people need those dollars to feed their kids (effectively, a gun to their kids’ heads), others’ dollars do have value #2: the dollar holders can coerce the dollar needers into providing something specific that they designate. i.e., working at some specific task. The quantity of this coercive power is a matter of degree, of course.
Put all the exchange relationships together, and we can say that a “representative dollar” carries some quantity of value #2: coercive power. Hence thneediness.
I might now have caught up to where your thinking begins.
“So coercers regulate their issue of coupons based on the size of the real economy and their perceived cost of coercion.”
This makes sense. I was trying to get to the political mechanics of how this plays out — how the currency of votes translates into currency, so to speak. The “cost” of coercion for an individual politician is not getting re-elected. And “coercion” for them means taxing more or spending less — issuing less net thneeds (running a budget surplus).
“Government debt does not represent or proxy for any sort of real investment.”
I guess this is right (though I’m vague on how it follows from the preceding), and if so my thought experiment doesn’t make sense.
“[ex nihilo dollar issuance] is valuable to the coercers, because they earn a “seigniorage rent†from the option-value associated with their coupons and a perpetual float associated with the public’s willingness to hold some stock of coupons in perpetuity, which allows them to obtain more resources for favored purposes through this scheme than they would have obtained through direct coercion to their friends. Plus, since a service (thneed provision) is attached to the coercion, coercion is much easier and therefore less costly to perform than would be naked confiscation on behalf of friends.”
So: the “market” does the coercion for them, at arms length? They don’t pay for the negative externality of ill will (by being voted out of office)?
@JKH
“the economics profession is beholden to the logic of accounting more than it appreciates at this stage”
I would rephrase: the economics profession *should be* far more attentive to the logic of accounting. And yes: I call MMTers the “accounting-based” school of ecnomics because they are attempting to do what economics in general “should” be doing. IMNSHO.
“Economics education programs should require rigorous training in the basics of financial accounting”
Yeah it astounds me: shouldn’t deconstruction/analysis of national accounts be required at about the 200 level, at least one semester?
I don’t know if I’m an MMTer. I certainly find many of their concepts and constructs to be intuitively appealing, coherent, convincing, and seemingly useful in trying to understand how economies work.
“So we assume real assets are held entirely by the private sector in this closed economy.”
Jumping ahead to your discussion of my critical error (equity versus equity claims) and returning to my “ownership” discussion in response to Steve Waldman, hoping this is not a confusion:
Entitites cannot “hold” real assets, only claims on them. Even if you’re holding an apple in your hand, your physical posession is not ownership; ownership consists only of your legal right to continue holding, exchange, destroy, or consume that apple. (This makes me think of, and makes sense of, the incomprehension we hear reported from indigenous peoples regarding land “ownership.”)
As you say, that legal right *is* a financial asset. Which means that my notion of disappearing all financial assets except bank money is a meaningless pipe dream. It would involve eradicating all claims on — rights to control/use — real assets. Those claims will always exist.
I suppose we could *define* “financial assets” as only those rights/claims that are so represented. If they’re not, they’re just rights. My right of free speech has not been monetized. Yet.
“the first step is that the government swaps $ 10 trillion of existing publically held debt for $ 10 trillion in bank deposits. This provides $ 10 trillion of additional bank deposits AT THE VERY START to”
Right, this is I think equivalent to the government having just issued bank credit in the past instead of debt, dollar bills instead of t-bills.
“Anyway, I’ll assume $ 10 trillion as a round number is closer to the cumulative deficit than $ 15 trillion.”
Seems about right to me, especially if you were to discount the circa $1.6 trillion in government debt currently held by the Fed, assuming the Fed is part of the consolidated government.
“The facilitating Fed/Treasury accounting arrangement for conversion is a detail”
Right: again, a consolidated entity.
I totally get the RHS equity thing, having been blessed with “possession” of some in various situations. (I do like the conceptual understanding that it’s only a measure.) But:
“The economy will have total assets of $ 50 trillion – $ 10 trillion in NFA and $ 40 trillion in real assets. This is offset by $ 50 trillion in RHS equity.”
That’s where something’s been missing for me, I think. After the quadruple-entry work has been done to consolidate national accounts from individual accounts, the current system doesn’t show any tally of RHS equity, at least labeled as such. Does it under another name? If so, where?
I’ve much liked Steve Waldman’s notion of citizens having “equity” in government, a discussion I think you’ve participated in quite a bit.
I need to stop now. Haven’t had the time to understand or respond as much as all your help deserves. I’ll definitely be coming back to these writings over days and weeks ahead.
@Asymptosis
Steve,
“the current system doesn’t show any tally of RHS equity, at least labeled as such. Does it under another name? If so, where?â€
Here:
http://www.federalreserve.gov/releases/z1/Current/z1.pdf
Page 106
Net worth
Currently $ 57 trillion
That’s household RHS equity
All financial claims on business and government transmit net to household balance sheets (except for international investment position leakage)
And all financial claims on business reflect a marked to market valuation of their physical assets
Nothing escapes ultimate valuation at the level of the household balance sheet
This is why the flow of funds statement (sources and uses of funds for micro) is so important
@JKH
“except for international investment position leakage”
i.e. the current US position with ROW is negative NFA, reflecting cumulative current account deficits and nominally matched gross flows, all marked to market
“Nothing escapes ultimate valuation”
i.e. NFA and all claims on the business sector are in there, all netted out for reverse direction claims issued by households (e.g. mortgages) and the international position
its all in there, in the $ 57 trillion
@vimothy
Vimothy,
“Not sure why this should be so controversial amongst a group that prides itself on the rigor of its sectoral analysis. For some reason, they seem allergic to the whole notion of public saving.â€
Exactly –
The reason is that it obscures the positive emphasis on NFA, which is the essential foundation for their approach to policy for deficit spending. The government’s position with non government is (NFA), and you don’t want that cluttering up the works.
This is the aspect that confuses some people about the definition of saving in MMT – there’s a reluctance to reconsolidate back to the national accounts definitions that generate the NFA definition in the first place, and that reconsolidation would help explain the full algebra of saving concepts (as I attempted to do above).
The positive emphasis on NFA is quite reasonable, but it’s not necessary to outlaw the normal consolidation of government and non government accounts in doing so.
So what is the net worth of the United States?
Do you look at household net worth (in the ballpark of $60 trillion).
Or total assets (approx $200T per John Rutledge, though I’m unclear what liabilities are offset against this).
http://rutledgecapital.com/2009/05/24/total-assets-of-the-us-economy-188-trillion-134xgdp/
Or since debts are paid by people and not assets– in personam and not in rem, as lawyers would say– perhaps we look at the human capital stock of the United States, estimated by a study from the Bureau of Economic Analysis study (“from the people who brought you the GDP”) at $750 trillion. Hell, with 2% inflation, that means the nominal value of US human capital is increasing $15 trillion a year. Its like we pay off the national debt in full every year.
http://webcache.googleusercontent.com/search?q=cache:XuMUsaSoCEsJ:www.bea.gov/scb/pdf/2010/06%2520June/0610_christian.pdf
The moral of the story, deficit hawks really are some of the worst humanity has to worry.
@beowulf
The Rutledge analysis is a complete joke.
He’s adding total financial assets to total real assets.
If the starting point includes total real assets, you have to net all financial assets to zero (horizontal), except for NFA (gov and foreign sources).
If you look at his numbers, that adjustment comes out ballpark close enough to the $ 57 trillion.
(The household net worth calculation includes household real assets, plus household NFA. That sector NFA includes all net claims on the other three sectors – business, gov, and foreign – and the claims on the business sector reflect the value of the real assets outside the household sector).
No comment on the human capital stock version, except have fun with it.
Right but Rutledge’s point (which he sort of backs in to) is that its is as asinine to measure national assets without netting out liabilities as its it is to measure national liabilities without netting out assets.
As for the human capital stock, like the poet said, man is the measure of all things. I’m simply giving you the conservative, lowball, official number from the BEA. If you were to measure it by how the EPA and OMB value a statistical life ($9 million) and multiply it by the 2010 census (309 million), the human capital stock equal $2.78 quadrillion. 2% inflation means nominal value increases by $55.6 trillion a year. I admit those numbers may be wrong. The Census Bureau may have overcounted the US population or, more likely, it could be that the EPA cost benefit analyses put far too great a value on human life (and as a consequence, our environmental and safety regulations are way too onerous).
Even if we were to drastically scale back the EPA value of human life by two-thirds (maybe rivers will catch fire again, like the good old days!), you’d still be north of $750 trillion. My point is, I’ve never seen the Nation’s human capital (by either the BEA’s measure or the EPA’s) netted against the Nation’s public debt, have you? Its like Thomas Edison said (and every instinct tells me Tesla would agree): :o)
“Look at it another way. If the Government issues bonds, the brokers will sell them. The bonds will be negotiable; they will be considered as gilt edged paper. Why? Because the government is behind them, but who is behind the Government? The people. Therefore it is the people who constitute the basis of Government credit…”
http://prosperityuk.com/2000/09/thomas-edison-on-government-created-debt-free-money/
@JKH
Okay so what sort of got me started on all this in the first place: how does individual saving/not spending in period 1 result in more “loanable funds” in period 2. Or does it?
@beowulf
Beowulf,
The general purpose of the accounting exercise is NOT to infer normative judgements about the relative balance sheet size of the debt. Fullwiler’s paper on fiscal sustainability is probably the best place to go for that. I prefer that approach to a gargantuan, meaningless present value calculation, offered as a sort of foil to a nominally large debt number.
Present value is fraught with pitfalls in any application. In that context, the Christian paper can’t be taken seriously, since most of the calculation consists of present valuing fictitious “lifetime nonmarket labour incomeâ€, for which there is no corresponding income or cash flow at all. It’s an interesting make believe exercise, but has nothing to do with the actual monetary economy.
There are probably better methods of present value than that. If forced, I suppose I’d start by capitalizing future NGDP, although with some significant qualifications as part of the interpretation.
Interest cost on the debt, being an income transfer, isn’t exactly a non-issue, even from an MMT perspective. Fullwiler deals with it. And Mosler recognizes it implicitly in proposing zero interest rates, in order to eliminate that type of income transfer. MMT has an enlightened approach on the subject, in that it tends to emphasize accrual accounting cash flows as opposed to placing full faith in present values. Scenarios for projected NGDP and the interest cost included in it, if done with reason, can capture an interpretation of the debt in an informed way, and certainly in a way that’s legitimate and superior to Rutledge/Christian methods.
There’s a host of ways of approaching the antidote for PPTDD (post and pre traumatic deficit disorder), but the Rutledge and Christian articles shouldn’t be in the mix.
It all begs for a “Ministry of Silly Calculationsâ€, fashioned on the template:
http://www.youtube.com/watch?v=b558kjihQQg
@Asymptosis
Steve,
“Okay so what sort of got me started on all this in the first place: how does individual saving/not spending in period 1 result in more “loanable funds†in period 2. Or does it?â€
It doesn’t.
(that it doesn’t, IMO, is pure Keynesian (inc. MMT), anti-Treasury view, anti-Says Law.)
Here’s one of the most constructive ideas from all of the MMT literature:
“Every transaction in a real-world economy affects financial statements of those engaged, and if an economic theory or a posited model is not consistent with how real-world financial statements are affected, then the theory is inapplicable.â€
(Scott Fullwiler)
That captures the fundamental dependence of economics on accounting.
It says that any conceivable, viable future real world economic scenario must be resolvable in terms of accounting at that future point. It’s a mathematical constraint on effective economic analysis.
In particular, it implies that any projected future macro income statement must balance, as per:
C + I + G + (X – M) = C + S + T
In a closed economy,
C + I + G = C + S + T
So, according to your example, assume case 1 of withholding income in the amount Z from spending on C (GOODS) and reallocating that same income Z to saving S.
That’s also an inventory build up where C (Z) stays on the store shelf, and becomes I (Z).
Is that more loanable funds?
No. There’s been no change in GDP in period 1. There’s been a reclassification of expenditure from C (Z) to I (Z). Z consumer goods “become†Z investment goods. And there’s been a reallocation of income Z from C (Z) (expenditure on C) to S (Z) (withholding of spending on C – i.e., saving S).
This can set off recession dynamics in future periods, as businesses adjust to inventory build up.
Now, assume case 2 of withholding spending in amount Z on C (SERVICES), and move that into S instead.
That’s also a change at the micro level. Some previous potential buyer is doing that saving within period 1.
Suppose the seller of that C service can’t find another buyer.
That means HIS income contracts.
It also means he has to borrow in order to buy the goods and services he bought otherwise. That means he dissaves at the margin the same amount saved by the first guy.
At the macro level, it means that the additional micro saving Z is offset by additional micro dissaving (Z). There’s no change in macro saving.
But the service C (Z) previously provided has literally disappeared, so GDP declines as a result.
Is that more loanable funds?
No. There’s been a reduction in GDP in period 1. C has declined, and S has remained the same.
The recession dynamic has already started in period 1.
P.S.
“It’s a mathematical constraint on effective economic analysis.”
i.e. a constraint on economic analysis, as a necessary condition for that economic analysis to be effective
(necessary, but obviously, not necessarily sufficient)
@beowulf
Beowolf,
You might like to have a look at,
Keith M. Carlson, The US Balanec Sheet: What Is It and What Does It Tell Us?”, Federal Reserve Bank of St Louis
research.stlouisfed.org/publications/review/91/09/Balance_Sep_Oct1991.pdf
This paper is 20 years old but still a good introduction to national income accounting and trends in US national wealth.
It also provides some estimates of national net wealth that include tangible and intangible human capital that you might find interesting, with cites of relevant papers.
@JKH
Thank you! That’s what I thought. I totally get what you’re saying in this comment, and can even condense it usefully:
Individual saving is simply leaving money sitting. It cannot possibly increase the quantity of money, financial assets, or “loanable funds.”
Here’s another question for you: could the tables in the NIPAs be rejiggered to remove the word/concept “saving,” without materially damaging (perhaps even improving) their modeling/conceptual value?
I’m hoping this question gives you the opportunity to drink more scotch.
Related: a comment I just posted over at Nick’s place:
@Asymptosis
Steve,
If you want to understand the neoclassical model, you need to ask a neoclassical!
The correct way to approach loanable funds is as a market for deferred consumption of real resources.
MMTers might have an accounting-consistent framework, but they lack a model of the real economy, which is why they don’t understand neoclassical econ in general and the market for loanable funds in particular.
JKH and Beowulf:
I very much like the MMT notion that there’s no such thing as government saving(s). If government issued money by spending and redeemed it with taxes, with no bonds issued, it would especially make sense. A bowling alley doesn’t “save” points.
But I also totally get how it’s an at least useful if not necessary accounting construct to usefully represent the national/world economy as it exists.
@Asymptosis
“Here’s another question for you: could the tables in the NIPAs be rejiggered to remove the word/concept “saving,†without materially damaging (perhaps even improving) their modeling/conceptual value?”
You haven’t seen me write on scotch, yet. You’ll know when that happens.
Let me come back on that particular question, though.
Pending that, this is what I was about to post before seeing your question:
BTW,
Accounting includes the “portfolio effect†of correctly interpreting income, balance sheet, and flow of funds statement for what they each add to the exercise.
An incorrect interpretation of the full set of statements may well lead to misguided efforts to revise one of them (e.g. NIPA), in search of the holy grail of monetary policy accounting. That’s neither necessary, nor advisable, nor achievable.
There is nothing in monetary policy that can’t be interpreted properly on the foundation of the existing set of accounts. As already noted, the flow of funds statement can be critically helpful in this regard.
@vimothy
To clarify, I don’t mean to suggest that JKH or Beowolf or any of your other commenters here lack understanding of anything in particular. Doubtless, if anyone lacks understanding, it is me. I just mean that as a general proposition, asking MMTers about neoclassical econ is like asking neoclassicals about MMT. You might get an answer, but it won’t necessarily be a fair one.
@vimothy “The correct way to approach loanable funds is as a market for deferred consumption of real resources.”
I understand that reasonably well. I totally get that there can be a market for loanable funds, assuming the stock of funds exists. (See Nick’s recent response to me on S/D diagrams for a fixed stock of antiques, which like funds are not produced or consumed.) But how does that explain this from Wikipedia?:
“Savers supply the loanable funds”
How does (not spending | buying/producing fixed assets) create loanable funds?
@vimothy “asking MMTers about neoclassical econ is like asking neoclassicals about MMT”
Right. I asked Nick the same question I just asked you. He suggested that he’s planning to post on the subject.
@Asymptosis
Steve,
Underneath it all, below all the financial intermediation, there are patterns of real resource consumption and deferral of consumption. That is where the action is from a loanable funds point of view, IMO. Think of income as output. Think of saving as deferred consumption of that income. Think of borrowing as a consumption of that income brought forward in time. You cannot print income. It must be produced.
P.S. to # 30
NIPA does not provide the full accounting framework that a monetarist analyst might want
But its not supposed to
That’s the reason for adding flow of funds and the balance sheet into the normal accounting mix
So there’s no reason to trash NIPA on that basis, or invent new stuff that’s already covered
One just needs to learn and understand what a flow of funds statement is, conceptually
BTW
The Fed flow of funds report is not “a” template flow of funds accounting statement, obviously.
It’s a whole bunch of statements cutting across all categories in the economy, most of which are sectoral flow of funds accounting statements, and some of which are balance sheets. There’s even a NIPA summary in there I think.
The standard flow of funds statement is most easily seen in corporate reports where it is typically referred to as the “sources and uses of funds” statement.
(Just clarifying; you probably know all this)
@JKH “So there’s no reason to trash NIPA on that basis, or invent new stuff that’s already covered. One just needs to learn and understand what a flow of funds statement is, conceptually”
This goes precisely to my thinking. I take your word for it that the current portfolio does everything that’s needed — IF you’re JKH, who has fully grasped and internalized the concepts and practices underlying that portfolio. As JKH says: “One just needs to learn and understand what a flow of funds statement is, conceptually.” I would add: “and its relationship to the other parts of the portfolio.”
I’m here to suggest that very few economists have done so. Like, almost none. I just searched Harvard and U Chicago econ course offerings/major requirements. The string “account” barely appears. Chicago says quite explicitly, “Courses such as accounting, investments, and entrepreneurship will not be considered for economics elective credit.” Much less requirements!
And I’m wondering if a rejiggering of the NIPAs to remove the manifestly confusing notion of aggregate “saving(s)” could help ameliorate the resulting widespread misunderstanding (which I’ve spent years trying to sort out in my head). My imagined tables would be no less consistent, and would mesh just as seamlessly with the the other parts of the portfolio. But they would make it easier to understand how economies work. In particular they’d get rid of that pesky S=I identity, because S would no longer exist. Could it be done? Usefully?
This is just pie in the sky thinking, of course, but it does go to the core of Nick’s recent post: we can’t really abolish “saving” if the NIPAs are lousy with the stuff.
Steve,
Returning to you question:
“Here’s another question for you: could the tables in the NIPAs be rejiggered to remove the word/concept “saving,†without materially damaging (perhaps even improving) their modeling/conceptual value?â€
My first thought here was to adopt the model used by “The Artist formerly known as Prince†and accordingly change the label “saving†to “The Residual formerly known as Savingâ€.
But seriously, folks.
If one has a problem with saving as a residual, realize that the connection of saving to something with more of a feel to it actually occurs at the interface between NIPA and the corresponding generic flow of funds statement.
Flow of funds identifies various NIPA saving components as sources of funds.
For example, in the case of households, it will match saving to incremental uses of funds, including bank deposit balances, bonds, stocks, real estate, pension funds, and on and on.
If you drill down, you can no doubt construct a flow of funds statement that includes the purchase of antiques as a use of funds.
And then you have to drill down to the individual household level to extract gross flows where money that may or may not have been sourced from current period saving is swapped inter-household for saving. Etc. etc. And on and on…
This is NOT a problem with the conceptual structure of NIPA saving.
The conceptual framework to capture antique purchases exists, as intended, in the flow of funds statement. It’s just a matter of developing the granularity in the data down to the desired level. That’s a data issue – not a conceptual accounting issue.
The end result being that you can wrap whatever monetary analysis model you want around this existing accounting framework in its full, intended scope.
But it helps to know what a flow of funds statement is, as part of the standard set of financial statements that must be interpreted jointly to tell a coherent story.
Let me suggest something, and be a bit blunt about it. I think Nick’s problem of being aggravated by NIPA limitations is that he’s not so familiar with what a (flow of funds)/(sources and uses of funds) accounting statement is and the nature of the information it provides at a conceptual level. I think Nick’s problem with NIPA saving has nothing to do with NIPA.
But in honour of Nick, I have constructed the following “ode to non-thingsâ€:
…………..
Tax in a closed monetary economy is a monetary measure of a subset of income.
It is a specific subset and allocation of income.
Income allocated to consumption is a monetary measure of a subset of income.
It is a specific subset and allocation of income.
Saving in a monetary economy is a monetary measure of a subset of income.
It is a residual subset of income with a variety of possible allocations.
One economist recently described the usual definition of saving as a “non-thingâ€.
If that’s the case, then income must also be a “non-thingâ€. A “non-thing†can’t be a subset of a “thingâ€.
And why not tax as well?
Or the income that ends up being allocated to consumption?
Does income have no separate existence prior to the event of purchasing consumer products? Is it not a “non-thing†during that existence?
“Non-thing†in this case effectively means “monetaryâ€.
You don’t resolve the dilemma of being assaulted by a monetary “non-thing†through attempting to make it a “thingâ€, such as antiques or bonds.
It’s not the residual nature of the allocation that makes saving a “non-thingâ€; it’s the monetary character of it. The other directly allocated components of income have the same monetary property, as income itself does, in a monetary economy.
Conversely, saving in a barter economy is a “thingâ€. It must be. Nothing is monetary in such a pretend economy. There are no “non-things†in such an economy.
………..
In conclusion, I see no need to rename “saving†to something else.
But something I haven’t delved into yet is that the whole “problem†of the residual as a sort of ontological thorn is that there MUST be an algebra of saving (and dissaving) across sectors and subsectors of the economy, below the purest level that exists only at the global level. And even at the global level, one must assume globally balanced government budgets (at least on average) to come out with global S = global I.
If you work out the logic of that multi-layered, sectoral divisibility requirement, you’ll find that there’s no alternative but to define saving as a residual.
@vimothy My intuition here — thinking that I haven’t worked out fully — is that there’s a confusion/confution here: of consumption with “consumption spending.”
We consume more each year than we spend on consumption spending — there’s all those fixed assets we’re consuming. That is subtracted from gross investment to derive net investment, but it’s not counted as current year “consumption.”
That’s as far as my thinking goes so far.
“Chicago says quite explicitly, “Courses such as accounting, investments, and entrepreneurship will not be considered for economics elective credit.†Much less requirements!”
F*!K, F*!K, F*!K, F*!K, F*!K, F*!K, F*!K, F*!K, F*!KING, F*!KERS!!!
@Asymptosis
I did my only course in financial accounting, and an entire course in flow of funds analysis, as part of an MBA.
Somebody was having a discussion on one of the blogs the other day, about doing a masters in economics.
I should have suggested doing a CFA.
The CFA program ranks accounting pari passu conceptually with economics, both positioned as part of a supporting framework for investment analysis.
@JKH
“”The Residual formerly known as Savingâ€.”
Love it.
“the connection of saving to something with more of a feel to it actually occurs at the interface between NIPA and the corresponding generic flow of funds statement.”
Yeah I was afraid of that.
“Flow of funds … match saving to incremental uses of funds, including bank deposit balances, bonds, stocks, real estate, pension funds, and on and on.”
Yeah, I got this from Kuznets. And *this* I like. Where I get lost is the apparent NIPA implication that all individual savings are instantly and perfectly intermediated into investment spending.
You get why I (and most economists, I think) think that’s what the NIPAs are saying?
“This is NOT a problem with the conceptual structure of NIPA saving.”
No: the problem with the NIPA structure (if a problem exists) is that it leads even most economists to misunderstand all this.
Are you suggesting that the current NIPA structure is the only possible, or best, structure? That it couldn’t be restructured in a more conceptually useful, easier to grasp form? I get that it’s not necessary for you, but…
“you can wrap whatever monetary analysis model you want around this existing accounting framework”
No: YOU can. 😉
“”ode to non-thingsâ€:”
Love it some more.
“Non-thing†in this case effectively means “monetaryâ€.
That’s what I was afraid of. We’re back to understanding the nature of money. I don’t think they teach that (very well, if at all) in econ school. Medium of exchange or unit of account? Floor wax or dessert topping? It’s both!
“You don’t resolve the dilemma of being assaulted by a monetary “non-thing†through attempting to make it a “thingâ€, such as antiques or bonds.”
This is why I felt the apple-economy thought experiment was less than useful — for me, at least.
“Conversely, saving in a barter economy is a “thingâ€. It must be. Nothing is monetary in such a pretend economy. There are no “non-things†in such an economy.”
Yes! If we or I save/store surplus wheat or drill-press-goodness we/I can consume it later. If I save/store surplus money (“we” can’t; the only surplus can come from government/international/aliens), I can’t. I think I get confused because “saving” means both of these things.
………..
“In conclusion, I see no need to rename “saving†to something else.”
I’m not suggesting renaming it; I’m suggesting eradicating the concept from the NIPAs entirely. Probably just a wacky notion.
“there MUST be an algebra of saving (and dissaving) across sectors and subsectors of the economy, below the purest level that exists only at the global level.”
I really respect your understandings; they are leagues beyond mine. But respectfully: are you really sure of this? I take your point that Nick probably doesn’t understand FOFs as clearly and intuitively as he does, for instance, IS/LM. But might his latest post be right, that the constructs we use (i.e. NIPAs with “saving”) make it almost impossible to think about these things in any other way?
@Asymptosis
thanks for your feedback!
“But something I haven’t delved into yet … there MUST be an algebra …are you really sure of this?”
I’ll take a break, and patch something together later today.
@Asymptosis “I’m suggesting eradicating the concept from the NIPAs entirely.”
cf Godley.
Steve: Let’s re-ask the question this way:
Suppose there’s an increased demand for financial assets by households (a rightward shift in the demand curve). Will that increased demand lead to an increased quantity of investment by firms and an increased quantity of financial assets sold to households (a movement along a supply curve)? It may do. That depends on the model. It’s a behavioural question, not an accounting question. It depends on the slopes of those demand and supply curves, etc.
I feel as an economics student that I ought to defend the profession here. So, one thing, disguised as two;
1, Why should economics students come away from graduate study with a good understanding of the nature of money as opposed to, say, applied industrial organization or asymptotic theory? What is that makes “money” both the pons asinorum and the philosopher’s stone of all knowledge?
2, From my perspective, MMT might have a good grip on accounting (I’m not in a position to judge), but it’s grip on economics does not always look assured. If there is a trade off, why should we weight the accounting side as opposed to the economics?
From my perspective, MMT might have a good grip on accounting (I’m not in a position to judge), but it’s grip on economics does not always look assured.
Astrologers feel exactly the same way about astronomers!
:o)
@beowulf
Ha! My Dad’s an astrophysicist actually, with a g-index of 13–but I’m pretty sure that he doesn’t know anything about astrology…
Steve,
Regarding the earlier question of NIPA, saving as a residual, and:
“There MUST be an algebra …are you really sure of this?â€
A model for NIPA and Flow of Funds accounting statements:
Assume an economy with a government, one firm B, two households H1 and H2, and a foreign sector:
NIPA
H1 income 300
Salary 200
Dividend 100
H1 expenditure
Consumer goods 100
H1 saving 200
B revenue 300
Sales to rest of the world 100
Sales to government 100
Sales to H1 100
B expenses
H1 salary 200
B profit 100
The entire NIPA report flows from this data.
Total private sector income is H1’s income of 300.
(H1’s income incorporates the dividend distribution of B’s total profit of 100 and classifies this as income, instead of B’s originating profit. An alternative treatment would be to treat B’s profit as income, and the dividend as a capital distribution to H1 rather than an income distribution. It doesn’t matter. The way presented here makes things simpler for illustration.)
Any attempt to include more than 300 in total NIPA income would constitute double counting of income in any conceivable case. (Feel free to test that proposition)
Accounting period GDP is 300, B’s output/revenue.
Total private sector saving is H1’s saving of 200.
Again, any attempt to include more than 200 in NIPA private sector saving would constitute double counting.
The macro sectoral allocation of private sector saving:
Current account surplus 100
Government budget deficit 100
Saving allocated to Investment 0 (investment is 0)
Saving is allocated entirely in the form of NFA (net financial assets).
In the Keynesian framework,
C = 100
I = 0
G = 100
X = 100
S = 200
T = 0
M = 0
I + G + (X – M) = S + T = S = 200
S = 200
Private sector saving = 200
Government deficit = 100
National Accounts saving = 100
Current Account surplus = 100
GDP = 300
FLOW OF FUNDS
H1 Source of Funds
Saving 200
H1 Use of Funds
Purchase of government bonds 100
Bank account 100
B Source of Funds
Profit 100
B Use of Funds
Dividend 100
Foreign Sector Source of Funds
Bank account (100)
Foreign Sector Use of Funds
Purchase from B 100
(I’ve assumed the foreign sector pays for its purchase from B by debit to a deposit account it holds with a domestic bank (i.e. in the home country of B). From an international flow of funds perspective, that represents a debit to the asset created by a prior gross capital inflow to the domestic country. That debit amounts to a capital account outflow. And that represents the domestic capital account deficit that funds the current account deficit of the foreign sector. The debit of course flows to a credit to B’s bank account.)
The Residual Nature of Saving
Recall:
H1 income 300
Salary 200
Dividend 100
H1 expenditure
Purchase 100
H1 saving 200
By construction, H1 saving is private sector saving, which is S in the Keynesian model.
S is determined as the residual of income less expenditure on C, the latter being H1’s purchase from B. (There are no taxes in this accounting period.) This is all reflected in NIPA.
The residual determination of saving can’t be decomposed to a unique combination of the income that was associated with its origination. For example, there’s absolutely no reason to say that saving of 200 consists of part salary of 150 and a part dividend of 50. And there’s no reason to suggest any other possible combination of those two. The residual calculation is what it is, in that sense.
The deeper meaning of the residual nature of saving is the following. It is inherent to double entry bookkeeping that the books must balance. The accounts are logically structured that way. And the way that they must balance forces some human calculation to be the residual one in the end. This case is fairly simple. Given the construction of the problem, we know that total income is 300. And we know due to the simplicity of the problem that we only have two categories for decomposition – consumption and saving. So in determining the required balancing for double entry bookkeeping, we have two choices. We can start with consumption and end with saving, with saving as the residual for the balancing of the books. Or we can start with saving and end with consumption, with consumption being the residual. It’s done the first way.
Why couldn’t it be done the second way, with saving identified directly, and consumption derived as the residual? A quick look confirms this is an impossible way to approach things from an accounting perspective.
In order to identify saving directly, and not as a residual, one would have to confirm a reliable institutional reference point for it. Bank accounts come to mind. So one could measure changes in bank accounts over the accounting period, and attempt to identify what has been saved as a result.
The problem is that the same bank accounts are used for a variety of financial transactions that affect those accounts, but that have no affect whatsoever on the amount that has already been saved. Such transactions change the asset composition of the portfolio into which saving is deployed. But they don’t change the measure of saving. So tracking bank accounts would be a completely misleading way of tracking saving in an attempt to identify saving directly rather than as a residual. The result would be chaos in attempting to construct aggregate saving from its constituent deployment out of bank accounts into real and financial assets. This confusion is compounded by the fact that there is much activity in real and financial assets that has nothing to do with saving, since it records trade in assets that existing prior to the current accounting period.
So saving is determined the first way, with consumption being determined first, and saving being determined as the residual.
And as far as the determination of the asset portfolio into which saving is ultimately deployed over a given accounting period – that is inherent in the information provided by the Flow of Funds statement (along with other things).
The Flow of Funds financial statement uses NIPA as one input. One of the constituent NIPA inputs to flow of funds is saving.
So NIPA generates saving as a residual calculation from income, and replicates that same residual as a source of funds in Flow of Funds.
It is only in the Flow of Funds report that NIPA saving can be linked to a corresponding use of funds.
Recall:
H1 Source of Funds
Saving 200
H1 Use of Funds
Purchase of government bonds 100
Bank account 100
In other words, the original NIPA residual saving calculation is preserved and transferred to the Flow of Funds report intact, as a source of funds. Saving is then linked to more specific financial or real assets only when the equivalence is established between sources and uses of funds, within the Flow of Funds report.
H2 and an Antique Transaction
We assume household H2 has zero income and expenditures.
It has no effect on GDP or NIPA.
Suppose Household H2 sells an antique (previous owned) to H1 for 100.
(Assume no commission is paid)
This antique transaction is not part of GDP; there is no income and no saving.
It is an asset swap, only captured in the flow of funds statement:
H2
Source of funds
Sale of antique 100
Use of funds
Bank account 100
(H1’s flow of funds statement is the inverse of this)
Conclusion
There’s no logical way to define saving other than a residual.
NIPA calculates saving as a residual income component, and sends it to the Flow of Funds statement for analysis of how that part of income gets used in the overall resolution of real and financial asset balance sheet accounting. That’s where we can see the financial and monetary flows that relate to transactions reflecting not only the deployment of saving, but gross financial flows that have nothing to do with current period GDP, income, or saving. The antique transaction is an example of that.
Assuming desired granularity in the composition of income statements, balance sheets, and flow of funds statements of all types (macro and micro), one move to any depth of monetary analysis desired. The answer to the problem of not finding the monetary effect of antique transactions in NIPA, for example, does not lie in attempting to reform the logical account structure of NIPA. Take the flashlight, and starting looking at related balance sheets and flow of funds statements, and their relationships to NIPA. That’s where the answer starts.
Steve,
Regarding the earlier question of NIPA, saving as a residual, and:
“There MUST be an algebra …are you really sure of this?â€
A model for NIPA and Flow of Funds accounting statements:
Assume an economy with a government, one firm B, two households H1 and H2, and a foreign sector:
NIPA
H1 income 300
Salary 200
Dividend 100
H1 expenditure
Consumer goods 100
H1 saving 200
B revenue 300
Sales to rest of the world 100
Sales to government 100
Sales to H1 100
B expenses
H1 salary 200
B profit 100
The entire NIPA report flows from this data.
Total private sector income is H1’s income of 300.
(H1’s income incorporates the dividend distribution of B’s total profit of 100 and classifies this as income, instead of B’s originating profit. An alternative treatment would be to treat B’s profit as income, and the dividend as a capital distribution to H1 rather than an income distribution. It doesn’t matter. The way presented here makes things simpler for illustration.)
Any attempt to include more than 300 in total NIPA income would constitute double counting of income in any conceivable case. (Feel free to test that proposition)
Accounting period GDP is 300, B’s output/revenue.
Total private sector saving is H1’s saving of 200.
Again, any attempt to include more than 200 in NIPA private sector saving would constitute double counting.
The macro sectoral allocation of private sector saving:
Current account surplus 100
Government budget deficit 100
Saving allocated to Investment 0 (investment is 0)
Saving is allocated entirely in the form of NFA (net financial assets).
In the Keynesian framework,
C = 100
I = 0
G = 100
X = 100
S = 200
T = 0
M = 0
I + G + (X – M) = S + T = S = 200
S = 200
Private sector saving = 200
Government deficit = 100
National Accounts saving = 100
Current Account surplus = 100
GDP = 300
FLOW OF FUNDS
H1 Source of Funds
Saving 200
H1 Use of Funds
Purchase of government bonds 100
Bank account 100
B Source of Funds
Profit 100
B Use of Funds
Dividend 100
Foreign Sector Source of Funds
Bank account (100)
Foreign Sector Use of Funds
Purchase from B 100
(I’ve assumed the foreign sector pays for its purchase from B by debit to a deposit account it holds with a domestic bank (i.e. in the home country of B). From an international flow of funds perspective, that represents a debit to the asset created by a prior gross capital inflow to the domestic country. That debit amounts to a capital account outflow. And that represents the domestic capital account deficit that funds the current account deficit of the foreign sector. The debit of course flows to a credit to B’s bank account.)
The Residual Nature of Saving
Recall:
H1 income 300
Salary 200
Dividend 100
H1 expenditure
Purchase 100
H1 saving 200
By construction, H1 saving is private sector saving, which is S in the Keynesian model.
S is determined as the residual of income less expenditure on C, the latter being H1’s purchase from B. (There are no taxes in this accounting period.) This is all reflected in NIPA.
The residual determination of saving can’t be decomposed to a unique combination of the income that was associated with its origination. For example, there’s absolutely no reason to say that saving of 200 consists of part salary of 150 and a part dividend of 50. And there’s no reason to suggest any other possible combination of those two. The residual calculation is what it is, in that sense.
The deeper meaning of the residual nature of saving is the following. It is inherent to double entry bookkeeping that the books must balance. The accounts are logically structured that way. And the way that they must balance forces some human calculation to be the residual one in the end. This case is fairly simple. Given the construction of the problem, we know that total income is 300. And we know due to the simplicity of the problem that we only have two categories for decomposition – consumption and saving. So in determining the required balancing for double entry bookkeeping, we have two choices. We can start with consumption and end with saving, with saving as the residual for the balancing of the books. Or we can start with saving and end with consumption, with consumption being the residual. It’s done the first way.
Why couldn’t it be done the second way, with saving identified directly, and consumption derived as the residual? A quick look confirms this is an impossible way to approach things from an accounting perspective.
In order to identify saving directly, and not as a residual, one would have to confirm a reliable institutional reference point for it. Bank accounts come to mind. So one could measure changes in bank accounts over the accounting period, and attempt to identify what has been saved as a result.
The problem is that the same bank accounts are used for a variety of financial transactions that affect those accounts, but that have no affect whatsoever on the amount that has already been saved. Such transactions change the asset composition of the portfolio into which saving is deployed. But they don’t change the measure of saving. So tracking bank accounts would be a completely misleading way of tracking saving in an attempt to identify saving directly rather than as a residual. The result would be chaos in attempting to construct aggregate saving from its constituent deployment out of bank accounts into real and financial assets. This confusion is compounded by the fact that there is much activity in real and financial assets that has nothing to do with saving, since it records trade in assets that existing prior to the current accounting period.
So saving is determined the first way, with consumption being determined first, and saving being determined as the residual.
And as far as the determination of the asset portfolio into which saving is ultimately deployed over a given accounting period – that is inherent in the information provided by the Flow of Funds statement (along with other things).
The Flow of Funds financial statement uses NIPA as one input. One of the constituent NIPA inputs to flow of funds is saving.
So NIPA generates saving as a residual calculation from income, and replicates that same residual as a source of funds in Flow of Funds.
It is only in the Flow of Funds report that NIPA saving can be linked to a corresponding use of funds.
Recall:
H1 Source of Funds
Saving 200
H1 Use of Funds
Purchase of government bonds 100
Bank account 100
In other words, the original NIPA residual saving calculation is preserved and transferred to the Flow of Funds report intact, as a source of funds. Saving is then linked to more specific financial or real assets only when the equivalence is established between sources and uses of funds, within the Flow of Funds report.
H2 and an Antique Transaction
We assume household H2 has zero income and expenditures.
It has no effect on GDP or NIPA.
Suppose Household H2 sells an antique (previous owned) to H1 for 100.
(Assume no commission is paid)
This antique transaction is not part of GDP; there is no income and no saving.
It is an asset swap, only captured in the flow of funds statement:
H2
Source of funds
Sale of antique 100
Use of funds
Bank account 100
(H1’s flow of funds statement is the inverse of this)
CONCLUSION
There’s no logical way to define saving other than a residual.
NIPA calculates saving as a residual income component, and sends it to the Flow of Funds statement for analysis of how that part of income gets used in the overall resolution of real and financial asset balance sheet accounting. That’s where we can see the financial and monetary flows that relate to transactions reflecting not only the deployment of saving, but gross financial flows that have nothing to do with current period GDP, income, or saving. The antique transaction is an example of that.
Assuming desired granularity in the composition of income statements, balance sheets, and flow of funds statements of all types (macro and micro), one move to any depth of monetary analysis desired. The answer to the problem of not finding the monetary effect of antique transactions in NIPA, for example, does not lie in attempting to reform the logical account structure of NIPA. Take the flashlight, and starting looking at related balance sheets and flow of funds statements, and their relationships to NIPA. That’s where the answer starts.
unintended comment duplication; 21:45 OK
SR,
Having gone through the exercise above, it occurs to me there may be a more streamlined example to demonstrate the general point at issue here:
Suppose a nation has no saving other than its current account surplus. Therefore, investment is zero. Suppose the saving S is reflected in the retained earnings of business B.
As a matter of operational finance, suppose B first leaves the amount S it generates as profit in its bank account. And suppose its bank, having that amount S as a deposit liability, offsets it by purchasing a foreign financial asset. That closes the loop for micro and micro in terms of the acquisition of financial assets that correspond to the current account surplus.
Then suppose B uses the amount S in its bank account to buy a variety of domestic financial assets from other domestic agents. These transactions are simply swaps of cash for other financial assets. There is no income effect, although such transactions will be recorded on the Flow of Funds statements for the various agents involved.
In addition, suppose B borrows an amount W domestically and buys an even greater variety of domestic financial assets with that amount W.
Suppose amount W = amount S.
So, among other things, B now has on its balance sheet:
Domestic Financial Assets
Amount Z (where amount Z = amount S + amount W = amount 2S)
Borrowing
Amount W (where amount W = amount S)
Retained Earnings
Amount S
And suppose Z consists of the following portfolio:
Z1
Z2
Z3
Z4
Z5
Z6
Z7
Z8
Z9
Z10
Where the sum of the Zi = Z
Now consider the problem of defining saving S in any other way than as the residual of income and expenditure – that residual being B’s profit and retained earnings in this case.
The only place you can go to try and find an “out†in respect of that challenge is to look at the asset side of where the S ultimately went.
And you CAN’T do it.
You CAN’T identify the subset of Zi that corresponds to S.
In real world operations, there will have been many individual transactions, temporary cash balances, etc. that go into the mix of the commingling of funding sources S and W, that eventually come out the other side as the commingling of funding uses across the Zi portfolio.
The detail of the AGGREGATE composition of the funds flow will show up in the Flow of Funds (Sources and Uses of Funds) Statement, but it will NOT show up as a segregation of S per se into corresponding uses of funds. There is simply too much operational granularity and commingling of the overall flow of funds to do it that way – even in a single example, let alone at the generic level of the NIPA income statement and corresponding macro flow of funds.
This sort of stuff is glaringly obvious if one is generally comfortable with the nature of double entry bookkeeping in the logical structure of financial accounting and how financial accounting tracks real world economic and financial activity. It’s all inherent in the accounting relationships among income statements, balance sheets, and flow of funds. And this is essential stuff to understanding the realistic interpretation and application of economic models that must de facto be overlaid on such a foundation of rigorous accounting logic.
Finally, in the logic of all of this, it is important to distinguish between S as a generic measurement concept or measure, and a particular amount labelled as S in a particular case as a particular application of that measure to monetary or real items in that case. The amount S may be used to acquire an asset, but that asset is never equal to S. The amounts are equal, not the substance of what each is referring to. From a financial accounting perspective, S always refers to RHS equity in this context, not LHS assets.
@JKH
Thanks. I really do appreciate your taking so much time. I’ve read twice (+) and I get it.
Maybe I’d be happy if “saving” was called “unspent portion of household income.” Or some such. Then I’d understand its relationship to undistributed business profits, for instance.
Oops. Getting to your latest, which goes straight to the issue of undistributed profits:
“As a matter of operational finance, suppose B first leaves the amount S it generates as profit in its bank account. And suppose its bank, having that amount S as a deposit liability, offsets it by purchasing a foreign financial asset.”
I could just leave it sitting in its reserve account, right? Or lend those reserves to another bank? But I get closing the loop.
“interpretation and application of economic models that must de facto be overlaid on such a foundation of rigorous accounting logic”
Ah, that it were so…
“The amount S may be used to acquire an asset, but that asset is never equal to S. The amounts are equal, not the substance of what each is referring to. From a financial accounting perspective, S always refers to RHS equity in this context, not LHS assets.”
*That* is great. I get it (mostly — not deeply or intuitively yet). IOW, as you said before, there is equity, and there are equity claims. I’ll keep revisiting this, hoping to imbue it into my conceptual bones. Thanks again.
“I could just leave it sitting in its reserve account, right? Or lend those reserves to another bank? But I get closing the loop.”
Yes. Excellent catch – I was wrong there, strictly speaking – oops.
The loop is closed by the reserve transfer alone, or at least what the reserve transfer reflects.
The reserve transfer clears a payment from a foreign buyer to a domestic supplier, which amounts to a marginal domestic current account surplus.
And the same payment reduces a foreign deposit balance with a domestic bank while increasing a domestic deposit balance with a domestic bank, which amounts to an international capital outflow (reduction of a prior inflow) as the “use†of the current account surplus funds.
“And suppose its bank, having that amount S as a deposit liability, offsets it by purchasing a foreign financial asset.â€
That’s effectively a gross capital outflow offset by a gross capital inflow. No effect on S per se. (And it leaves out a potential preceding FX transaction, which is also a pair of matching gross capital flows.)
Duh on me (think I’ll forgive myself just this once though).
Thanks for feedback and the opportunity to write this out here. It really helps with the learning process.
(If you don’t mind, I may return later just to try and put Steve Waldman’s very interesting comment within my own context that followed.)
@JKH
“That’s effectively a gross capital outflow offset by a gross capital inflow. No effect on S per se. (And it leaves out a potential preceding FX transaction, which is also a pair of matching gross capital flows.)”
And thank you, that answers another thing I wondered about: the effect of international “capital” flows on NFAs: nada.
“Thanks for feedback and the opportunity to write this out here. It really helps with the learning process.”
I’m glad that’s true for you as well.
“(If you don’t mind, I may return later just to try and put Steve Waldman’s very interesting comment within my own context that followed.)”
I’m thinking you’ll do a better job of grasping it and responding than I did…
Just a reminder: RSJ wrote these two posts on nominal vs. real a while back that have parallels with this discussion, (I think):
http://windyanabasis.wordpress.com/2011/03/31/nominal-versus-real-part-1/
http://windyanabasis.wordpress.com/2011/04/07/monetary-versus-real-part-2-why-monetary-exchange/
Having a clear concept of both and clearly distinguising them linguistically would seem a minimal requirement for any decent analysis. Using one as a proxy to describe the other does not qualify, in my mind. But the ultimate goal must surely be to bring them together, to describe the interface, through some sort of a philosophical translation algorhythm, in a coherent fashion? RSJ’s use of time I see as a proxy for such a logical merger, at least wrt changes from one state to another, but it isn’t an explanation in itself. Another similar proxy would be ‘uncertainty’. But then, maybe, this isn’t what economics was designed to do in the first place and it probably isn’t an attainable goal anyway, especially considering it seems most economists do not even qualify for the minimal requirements as stated above.
JKH: “Anyway, the end result of all of this is the following:
The economy will have total assets of $ 50 trillion – $ 10 trillion in NFA and $ 40 trillion in real assets. This is offset by $ 50 trillion in RHS equity.”
So, the answer to Steve’s original question: “Would the remaining money in all the bank accounts equal U.S. government debt?”
Is, “yes.”
As I understand it, another way to put this is: The value of all real assets is contained within all the financial statements of the economy, so netting out private debt to zero will leave real assets plus government liabilities.” ??
@ JKH, I am confused by your statement on national savings above (maybe because all my economics has been through MMT) : “The corresponding “saving†numbers are those generated per accounting period. For example, suppose the government runs a $ 1 trillion deficit and the economy adds $ 2 trillion in real assets, both in a given year. Then incremental NFA is $ 1 trillion, incremental investment is $ 2 trillion, incremental private sector saving is $ 3 trillion, and incremental national accounts saving is $ 2 trillion.”
In a closed economy, if investment is 2 trillion, isn’t that just dis-saving of the entity doing the investing? I thought this all needed to net to zero. Why net out public investment from national savings but not private investment?
@Dave
Dave,
Assume a closed economy with 3 sectors – business, households, and government.
(MMT combines business and households as the private sector.)
Two stages of analysis – private sector activity; government activity
Assume real investment of $ 2 trillion.
For simplification, business accounts for the entire $ 2 trillion.
Suppose business has financed the investment with $ 1 trillion debt, $ 1 trillion new common stock issue.
Payments go to factors of production that created the investment.
Assume those payments ULTIMATELY go to households in the form of wages etc.
That creates $ 2 trillion in income for households.
That income is saving.
That saving cannot be spent out of existence. It’s an accounting fact at this point.
It could be spent in the sense of a multiplier effect, but for simplification, assume households don’t spend and aren’t taxed any further; i.e. forget about knock-on noise like multipliers, etc. Treat this as a one-time $ 2 trillion injection of income and saving.
Suppose households acquire the $ 2 trillion in stocks and bonds issued by business. That is certainly possible, because it doesn’t change the fact of saving having occurred.
Accordingly, household balance sheets show an increase of $ 2 trillion in savings (i.e. cumulative saving, net worth or equity), and $ 2 trillion in financial assets.
Note the distinction here between (common) stock as a financial claim held by households, versus equity as a right hand side balance sheet entry for them. Households do not issue financial claims directly against their RHS equity.
(We still haven’t introduced government yet.)
Now, consolidate the private sector balance sheet, combining business and households.
Bonds and stocks are financial claims issued by business, also held as corresponding assets by households. The net effect from consolidation of claims issued and the same claims held as assets held is zero net financial assets. MMT describes this as “horizontal consolidationâ€.
What’s left after consolidation is a private sector balance sheet of $ 2 trillion in investment assets, and $ 2 trillion in private sector savings (i.e. cumulative saving, net worth or equity).
MMT tends to leave this $ 2 trillion gross nominal balance sheet equivalence to the side. MMT is more interested in the concept of net financial assets, which doesn’t appear yet in this balance sheet. Net financial assets so far are zero.
NOW, suppose the government runs a deficit of $ 1 trillion.
It purchases private sector output of $ 1 trillion.
Similar to the case of investment, and for simplicity of illustration, set aside any complication from noise like multipliers, crowding out, further taxes, etc. Assume that the $ 1 trillion expenditure by government results in $ 1 trillion additional income to private sector factors of production for the goods and services sold to the government. That income becomes saving, since by assumption all spending has already been accounted for (the new government spending, as well as previous spending by all sectors). Furthermore, assume the private sector purchases $ 1 trillion in bonds issued by government. (This bond issue could be viewed in the MMT sense as the “mop-up†of deficit spending.)
So the private sector balance sheet increases by $ 1 trillion in government debt assets and $ 1 trillion in savings (cumulative saving, net worth or equity).
This is what MMT refers to as a $ 1 trillion net financial asset position (NFA).
Combining the effect of $ 1 trillion in government spending with $ 2 trillion in private sector investment, the overall result is:
$ 2 trillion real investment (flow and stock)
$ 1 trillion NFA (flow and stock)
$ 3 trillion private sector GROSS saving (flow), with corresponding stock (cumulative saving, net worth, or equity)
$ 2 trillion national accounts saving (flow and cumulative flow as stock)
S = $ 3 trillion
NFA = $ 1 trillion
National accounts saving = S – NFA = S – (G – T)
Steve Roth / Steve Waldman,
This is my interpretation of Steve Waldman’s model, as per his comment. My iterative construction of Steve’s model may differ somewhat from Steve’s own presentation.
Assume an economy with two sectors – government and private. The conventional monetary system has been replaced by something else. All agents have liquidated their monetary positions along the lines of Steve R.’s specification and roughly as per Steve W.’s description.
The previous monetary unit is replaced by “thneed†currency. A “thneed†is a claim on real resources – some combination of real goods, services, and assets. The precise specification remains to be worked out.
I will refer to the “thneed†currency as THN.
The government can run balanced, deficit, or surplus budgets. It can issue debt. All government spending, taxes, and debt are denominated in THN.
The act of government taxation is called COERCION. (Note – Steve introduced the idea of coercion as something associated with coercion certificates and redemption of same – or debt redemption. But debt redemption on a net basis requires taxes. So I’ve generalized coercion to mean taxation per se.
THN is a claim on real resources. These claims have a certain optionality structure to them.
When the government coerces THN taxes, it demands specifically tailored THN from taxpayers.
The government is effectively long a THN call option; the taxpayer is short.
(I’m not 100 per cent sure Steve would agree with my option interpretation. I’m not entirely clear on what he intended.)
The taxpayer, being short THN tax liabilities, must come up with the real resource bundle that the government demands. The taxpayer may have little choice but to surrender a piece of his house or i-phone.
Similarly, on the expenditure side, the government spends THN by delivering it to other taxpayers, as per their general requirements/desires. Taxpayers on that side are long the call as a collective force, through their government, although some individuals may not be satisfied with the result.
So that’s the case with the regular flow of taxes and spending.
This government can also run deficits or surpluses.
When it runs deficits, it issues THN debt claims as assets to taxpayers. Here, the taxpayer has the call option, in the sense that he can determine the THN real bundle he is willing to deliver up front in exchange for coercion certificates redeemable in the future. The government then distributes THN debt issuance proceeds as part of its overall THN expenditures.
THN debt is a promise to repay THN in the future. When that happens, the government must either roll the debt, or pay it down with THN taxes. In the latter case, the government is in the position of coercing taxpayers through its THN call to come up with the real resources that the debt holders demand on repayment. This seems to be the dynamic that Steve emphasized in his piece.
I don’t know whether Steve intentionally shaped some of his interpretation from MMT, but I see certain similarities. Demand for THN debt is in effect a desire for “net THN assetsâ€, an analogue to net financial assets in MMT, with similar motivation for desired saving as the driving factor.
Steve W.’s piece responds to Steve R.’s original concern about the relationship between government debt and fixed assets. He says there is no direct relationship, and I agree, although that point seems muted within his overall presentation. I interpret the main issue to be the relationship between real investment and various classifications for saving. Steve W. answered with his THN model. I answered with my own outline of comprehensive double entry accounting, with a coherent set of financial statements.
(Steve R. mentions the issue of confusion about the scope of real investment in the sense of its coverage beyond “fixed assetsâ€, etc. The incorporation of the government’s own assets (as opposed to private sector investment) is part of this as well, I think. I’ll leave that. Let’s just assume that the system identifies correctly what constitutes real investment. My working assumption is always that this refers to private sector real investment, as is the intention in Keynesian type modelling.)
From my perspective, the important point here is that there is no direct connection between net financial assets in the form of government debt (in dollars or THN) and real investment. Moreover, there is a meaningful and important distinction between real investment and any defined class of saving. Saving measures can be constructed at all levels of the global economy – total, country, sector, sub-sector, or agent. It is NEVER true that saving is the same thing as investment, whatever the chosen tier of saving. It MAY be true that the quantity of saving equals the quantity of investment, depending on the classification of saving, the sector in question, and the actual measures recorded for each in that sector.
Wow, I’m absent from the thread a few days and I miss JKH redraft Macro.
Savings is a residual reminded me a point that Godley made
The only area where, in fact, supply and demand perfectly met was in the financial part of the economy. Everywhere else both individuals and firms needed what Godley called a “buffer”, a saved resource for future income and spending that whose price was itself uncertain. Historically, for individuals the buffer was saved money that did not depend on the government’s fiat; for firms it was money, in part, but also something more. For a goods producer the buffer was inventory; for a service firm it is slack labor capacity. That was, in fact, how the system of the production, distribution and sale of goods and services actually worked. In Godley’s words, “outside financial markets there is neither need nor place for equilibrium conditions to bring supply into equivalence with demand.”
http://www.siliconinvestor.com/readreplies.aspx?msgid=27682820
@beowulf “Wow, I’m absent from the thread a few days and I miss JKH redraft Macro.”
Yeah! Why doesn’t that guy have a blog? (JKH?) His work is seriously underutilized here in the Asymptosis comments section. (But aren’t we a lucky bunch.)
“Savings is a residual reminded me a point that Godley made
The only area where, in fact, supply and demand perfectly met was in the financial part of the economy.”
Aha moment! Perhaps related: I’ve been thinking that in fact the market for financial assets is the only part of the market that truly is a barter economy. I trade my checking account deposits for your stocks or bonds. I haven’t figured out yet how to characterize the real-goods market by contrast.
Great link. I wish Niederhoffer gave better links to the thinking he’s referencing. Have to do some googling…
@Asymptosis I see that that linked-linked post was not by Niederhoffer, but by Stefan Jovanovich. Who is *that* guy?