An Inflation Indicator to Watch, Part 2

“Inflation comes from too much money chasing too few goods.”
—Many an economics professor

But does it really?

In Part 1 of this series, I said I would challenge popular beliefs about inflation, while at the same time proposing an alternative theory. I also promised that the theory leads to an inflation indicator with an excellent prediction record, and I’ll define the indicator in this article. Considering inflation determines bull and bear markets in both stocks and bonds, that’s my big pitch.

But first, here’s a recap.

In the last article, I argued we should stop fixating on M2, as economists do, and focus instead on the amount of money banks pump into the economy when they deliver loan proceeds. Bank-created money, which I call “M63,” flows directly into GDP, thanks to the charters that allow banks to produce money from nothing when they make loans or buy securities.

Here’s a chart from the earlier article:

inflation 1

The chart shows bank loans lifting spending above the amount spenders can fund from prior income, which fattens the circular flow, boosting GDP. Bank-funded spending might lift real GDP or inflation, or both, but however it works into the economy, we can find a correlation with nominal GDP. And we can demonstrate the correlation with either loan or money data, as long as we define money as bank-created money (M63), not M2.

God’s A-Gonna Trouble the Water

You might wonder, though, if bank-created money outperforms M2, why do mainstream economists rarely discuss it?

To restate a few more points from Part 1, the answer lies partly in the backstory of the 1960s and 1970s, when the economics gods pulled a bait-and-switch. The gods baited Milton Friedman and Anna Schwartz with ninety-four years of M63 data, and Friedman and Schwartz then analyzed and wrote up the data in their magnum opus, A Monetary History of the United States, 1867-1960. Within ten years, the book was a classic, and Monetarism had become a key building block in mainstream economics. Thanks to M63, Friedman and Schwartz reached their profession’s pinnacle while becoming household names outside of it.

But somewhere along the way, the gods pulled their switch.

Instead of continuing to channel M63, they directed economists toward measures such as M2, which better matched the profession’s long-time fascination with the intrinsic characteristics of money (such as liquidity, stability and value as a medium of exchange). Economists were well-accustomed to treating money as an independent force that’s mostly unrelated to bank lending, and a close look at ninety-four years of history wasn’t powerful enough to break old habits. In fact, most economists were unaware of the differences between the traditional money measures that dominated their discussions and the M63 measure that Friedman and Schwartz studied, and they might still be unaware. Even Friedman and Schwartz gave no indication they considered the differences significant.

To be sure, the bait-and-switch wasn’t a big deal at first, because the various money measures were highly correlated. But the correlations soon weakened, and only M63 retained a strong correlation with economic growth. In other words, up-to-date research reaffirms Friedman and Schwartz’s original conclusions about M63 while refuting the supposed value of M2, adding to the ample evidence that M2 failed as an economic indicator.

Of course, that brief history of Monetarism isn’t well-known, if known at all outside a few small tribes of economics rebels. But the more you delve outside the mainstream, the more obvious it becomes that it’s the only history that matches real events. Spend enough time running with the rebels, and you’ll come to appreciate three facts that mainstreamers typically sweep aside or deny:

  1. Bank charters authorize commercial banks to create money from nothing when they extend loans, making bank lending fundamentally different than other types of lending.
  2. We can measure the amount of money that banks create.
  3. Bank-created money is often uncorrelated with the money measures you hear about in the media, such as M2.

Once you accept those three facts, you only need to look at the connections between the various money measures and the economy (see my article “Learning from the 1980s”) to confirm the case for refocusing on M63. And as noted above, it also helps to know that Friedman and Schwartz produced their seminal research using M63, not M2, for which data didn’t exist over the full period of study. Plot twists like that almost make the history of economic thought entertaining, as if written by Stephen King or M. Night Shyamalan.

There’s a Leak in This Old Economy

Needless to say, M63 feeds into the alternative inflation theory, but it’s not the only input. Let’s complete the theory by adding other inputs, and the next one also drops from the circular flow. It accounts for exports and imports, which I’ve added to the flow in the chart below (leaving out bank loans this time to keep it simple):

inflation 3

The chart shows exports boosting the income accruing to domestic producers (at the bottom), whereas imports have the opposite effect of channeling business to foreigners at domestic producers’ expense (at the top). Pretty basic, right? Maybe not especially interesting? In fact, foreign trade becomes interesting within the circular flow only when exports and imports fail to balance, as in the United States. Because of America’s trade imbalance, her circular flow doesn’t self-replenish, as shown in the next chart:

inflation 4

Notice that domestic income is shrinking with respect to spending, which is the piece I find interesting—that’s what I mean when I say the circular flow doesn’t self-replenish. Or, you can think of the flow springing a leak, whereby purchasing power leaks from the domestic economy with potential consequences for inflation. Before weighing those consequences, though, we need to consider the bigger picture.

In particular, we need to consider how foreigners use the dollars they accumulate when they sell more stuff to us than we sell to them. They have excess dollars they need to invest, thanks to the trade imbalance, but how? Are they investing their dollars directly into our “real” economy, or do those dollars first flow into our “financial” economy? Economists often downplay that difference, noting that the two economies connect, but my experience as an everyday observer, analyst and investor tells me it’s important.

If foreigners use their dollars to build a factory in, say, South Carolina, our real economy gets a boost that neutralizes the circular flow’s leak. But if they only buy stocks and bonds in our financial economy, that shouldn’t provide as much of an offset to the leak. (If you’re used to talking about the velocity of money, another way to make the same argument is to say that the circular flow slows down—meaning lower velocity—when purchasing power has to take a detour through financial markets.)

So which is it?

Well, data show the foreign sector mostly draining dollars from our real economy and then investing them in our financial economy. The real-economy drain (or leak) reduces pricing power and dampens inflation, while financial-economy investments boost asset prices. For this article, we’ll consider only the inflation effect. (I may write later about whether trade imbalances help U.S. stocks trade at higher multiples than other markets, which seems a reasonable secondary conclusion.)

Two Words You’ll Be Hearin’

Essentially, America’s trade deficit is deflationary, and you probably already believed that without me trying to convince you. I’ve supported the case using the charts above to show similarities between foreign trade and bank-created money—namely, both affect inflation by disturbing the circular flow. Once we connect inflation to the circular flow, we can see that

  1. Inflation should rise when purchasing power injected into the spending flow causes it to exceed the amount of domestic income from the prior period. The most common example, which I discussed in Part 1 and recapped above, is a surge in money-creating bank loans.
  2. Inflation should fall when purchasing power leaking from the spending flow causes it to undershoot the amount of prior-period domestic income. The most common example is a trade deficit that redirects the spending flow from domestic to foreign producers, meaning it no longer circles back cleanly and directly to the real economy, as discussed above.

More simply, inflation forecasters should be aware of injections to and leaks from the spending flow. Or even more simply, just think injections and leaks. In my opinion, those two words should appear in the job description of every would-be macroeconomist.

But that’s not all, because I haven’t yet accounted for the supply side, which could expand or contract to absorb the injections and leaks. Depending on the supply side, injections and leaks could have greater effects on real growth than on inflation. We can use a reasonable method for isolating inflation, though—we just compare the spending injections and leaks to real GDP growth, and that approximates whether spending is overshooting or undershooting supply.

All Together Now (Bom Bom Bompa Bom)

So my suggested inflation indicator is easy to calculate—we add M63 growth to the trade balance (as a percent of GDP) and then subtract real GDP growth. M63 growth and the trade balance measure spending injections and leaks (have I mentioned those terms?), whereas real GDP growth tells us about the supply side.

Conceptually, the indicator modifies the saying that inflation comes from too much money chasing too few goods. Instead of fixating on money, we’d like to know whether too much purchasing power is chasing too few goods, or, in the case of disinflation, whether too little purchasing power is chasing too many goods. The difference between money and purchasing power may sound trivial, but consider the following:

  • When Milton Friedman predicted high inflation in the 1980s (see Part 1), he cited strong M2 growth as the reason for his prediction. Had he allowed that purchasing power was leaking from the spending flow as America’s trade deficit soared, he may not have sounded his errant inflation alarm.
  • Also in the 1980s, M2 lost its strong correlation to the amount of purchasing power created by commercial banks. It no longer gave an accurate reading on the amount banks were injecting into the spending flow, and I would say it’s no coincidence that M2 lost its effectiveness as a leading indicator at the same time.
  • Data shows quantitative easing (QE) failing to inject much purchasing power into the spending flow, confounding those who expected the money created through QE to lead to high inflation. (See my article “QE’s Untold Story” for the “argyle chart” that shows how banks responded to QE.)

Clearly, money and purchasing power can be different, and I’m suggesting the better way to understand inflation is to look at purchasing power within the circular flow, not money.

You might think of a tug-of-war between purchasing power, on one hand, and capacity, on the other. The tug-of-war’s outcome depends on the circular flow’s particular pattern, which depends on spending injections and leaks, and that’s the crux of the alternative theory I’m proposing. To my knowledge, you won’t read the same theory in any other blog, book or paper, but that’s not to say it doesn’t build on the history of thought. You can find the theory’s origins in my recently published book, Economics for Independent Thinkers, which covers the circular-flow approach in more detail.

I haven’t yet shown the theory works, though, and I’ll do that in Part 3. As already promised, I’ll show that the circular-flow indicator has an excellent historical record.


  1. Andre


    Dear Daniel, I’m so glad I’m stumbled upon your writing. I think the world will beat a path to your door as soon as they come to the realization that bank injected credit matters (and understand it really is created out of thin air!). So far, Steve Keen is the only other economist I’ve come across that talks about this. Are there others of any note that have come to this realization yet?

    As I’m an engineer by trade, and not an economist (and thus actually open to new ideas!), I was wondering if you could spell out the main reason that M2 diverges from M63. After all, U.S. bank created credit always first appears in checking accounts, which is covered by M2. Perhaps you’re referring to the concept that when there is a ‘leak’ due to imports exceeding exports, the funds that would temporarily appear in savings accounts would then perhaps enter the “Euro Dollar” system inside foreign banks which hold Dollar balances that are unregulated by the Fed and where reporting is not easily possible and which does not appear in M2? Thanks!

    • ffw


      Thank you for your kind comments, Andre. I actually discuss the history of thought on bank-injected credit in my book – many others considered it important going back about as far as you can go – David Hume and David Ricardo, for example, opposed bank-created money, and it plays a role in the economics of Wicksell, Schumpeter, Minsky and the Austrian School.

      As far as M63, I include every type of bank deposit regardless of liquidity and term (unlike M2), because the change in total deposits closely approximates the amount of money banks are injecting into the economy. So if a bank extends a loan via a checkable deposit that eventually morphs into a long-term time deposit (after passing from one party to another and so on as people pay for things), it still injected money into the economy, regardless of how that money changes form on bank balance sheets, or at least as long as it remains on bank balance sheets. In other words, M63 growth is the amount that banks expanded total purchasing power from one period to the next, and that’s not affected by how deposits break down into maturity and liquidity buckets.

      I also subtract reserves, though, because deposit creation that occurs via open market operations typically isn’t initiated by commercial banks and is less likely to flow directly into the real economy. Over the period the Fed has existed, that makes for a difference with the Friedman and Schwartz measure, but it’s a very small difference over their period of study. I explained that further in “Learning from the 1980s.”

      I hope that’s more clear – I didn’t want to get into equations in the main text but it’s good that you asked.

      • ffw


        Forgot to mention – there are no money market mutual funds in M63, either, because money market funds, like a bond fund, transfer purchasing power from one party to another without increasing the total.

        • Andre


          Daniel, thank you so much for your detailed reply! A couple of things: I was a little confused about what you meant by “subtract” reserves, because I thought that electronic central bank reserves (even those injected via the discount window or OMO) never showed up in M2 anyway, and never actually appears as a deposit in anyone’s bank account unless the bank decides to create create that matches the reserve injection. Maybe I’m overthinking this and you simply are clarifying that central bank reserves don’t matter in terms of the money supply in the economy, which makes perfect sense.

          The other thing was whether you think that the “EuroDollar” system messes with our understanding of the money supply in large and unpredictable ways which can throw off an analysis. If I understand correctly, decades ago the Fed stopped trying to set interest rates based on the money supply, because it realized that the EuroDollar system was too opaque for them to understand. The extent of my knowledge about the EuroDollar system is based on these podcasts:

  2. Pingback: A Key Inflation Indicator To Watch, Part 2 | peoples trust toronto

  3. Pingback: A Key Inflation Indicator To Watch, Part 2 – iftttwall

  4. Pingback: A Key Inflation Indicator To Watch, Part 2 – ProTradingResearch

  5. Pingback: A Key Inflation Indicator To Watch, Part 2 – TradingCheatSheet

  6. Pingback: A Key Inflation Indicator To Watch, Part 2 | Real Patriot News

  7. Pingback: A Key Inflation Indicator To Watch, Part 2 | CENSORED.TODAY

  8. Pingback: A Key Inflation Indicator To Watch, Part 2 – Wall Street Karma

  9. Pingback: A Key Inflation Indicator To Watch, Part 2 - Investing Matters

  10. Pingback: A Key Inflation Indicator To Watch, Part 2 | Investing Daily News

  11. Pingback: A Key Inflation Indicator To Watch, Part 2 – The Deplorable Patriots

  12. Reply

    Your theory fits exactly with my own musings and the book I have written about the global financial system (unpublished as yet) encompasses some of your points, although you have admirably gone much further. Execellent read and well thought out. I should be happy to email you an electronic copy of my book if you are interested as any feedback is welcome.
    Best regards.

    • ffw


      Thank you, Peter, I appreciate your comment – always glad to learn of more people who think as I do. I’ll get back to you offline.

  13. Brian H


    I think your theory closely resembles and has many things in common with Keith Weiner’s theory on interest rates and prices. You are obviously both trying to bring the reality of credit to the flawed monetarist view. I find Keith’s explanations to be less intuitive than yours, but I think he also casts a much wider net by addressing irredeemable currency as the foundational underpinning for the ever-expanding credit.
    I’m really hoping you will know each others’ work, already… But just in case

    • ffw


      This is helpful, Brian, thanks! I know of Keith Weiner as a classical gold standard advocate – I saved one of his articles a few years back – but I don’t remember reading anything from him on bank credit specifically. I’ll definitely take a look.

  14. ffw


    Thanks Andre (replying to your reply further up the thread), I think there’s a lot of misinformation on the web about how OMOs work. Treasuries that the Fed purchases are mostly, ultimately sourced from the primary dealers who run the Treasury market. Those BDs can’t be paid via reserves because they’re not commercial banks – they don’t hold reserve balances at the Fed. (Most have sister companies with commercial banking charters but the BDs themselves are separate.) So the Fed pay for the bonds by increasing the reserve balances at the BD’s bank and then instructing the BD’s bank to credit the BD with a deposit. That’s the process that the Fed (and Bank of England) have long described for OMOs and they continue to describe it like that on the Fed’s website. That said, M2 doesn’t necessarily jump with OMOs for other reasons, and I gave my views on that in an article from last October, “QE’s Untold Story.” Here’s a Seeking Alpha version of that article that also includes a note at the bottom with links to sources of info on how OMOs work (I included the links because Seeking Alpha has promoted at least a few QE commentaries that get the mechanics wrong – they contradict processes described by the central banks themselves): “Why QE Is Overrated”

    Thanks for the link to Macrovoices, and I agree with your point – where I argued that the excess dollars held by foreigners don’t cleanly flow back to America’s real economy, I agree that eurodollars are one piece of that, because they soak up a portion of the foreigners’ excess dollars without any clear impact on economic activity.

    • Andre


      Aha, thank you for clearing this up. I had always assumed that the reason that the central banks used only primary dealers to engage in OMOs was that it was only the primary dealers that had both government bonds to sell and an account at the central bank to receive funds to. You’ve cleared up this misconception for me.

      I’m still, however, confused about why you’re subtracting central bank reserve injections from M63. I would have assumed that the BD would park the newly created deposits matching those reserve injections into the stock market, thus creating a link between asset price inflation and central bank creation of reserves. Perhaps you’re saying that the wealth effect caused by this asset price inflation is not an important factor in your model? Or maybe instead that the BDs aren’t actually putting these newly created deposits into the stock market?

      BTW, I’m in the UK and I noticed it’s not possible to order your book from to an international address because Amazon says the seller has restricted this ability. For anyone else reading this that had the same problem, I luckily found a link at the bottom of that let me order directly from the publisher. I look forward to reading it 🙂

      • ffw


        Thanks Andre, and thank you also for the book order and plug.

        As far as your comments on what BDs would do with their QE sales proceeds, I’m not sure why they would use their extra deposits to increase inventories of stocks, in particular, as against buying more bonds, but I agree that the question of how BDs manage their balance sheets through QE periods is relevant—hence this research I wrote up in October. I combined bank and BD balance sheets and found that the aggregate balance sheet grew just as quickly in between QEs as during QEs – QE appeared to have no effect at all on balance sheet growth. Some of that result was probably explained by banks/BDs frontrunning QEs, but that’s likely only a small piece of it – in my opinion, the balance sheet data is important to showing what QE really does.

        All that said, I don’t think the balance sheet data means that QE had no effect on asset prices, just that the asset price effects may be more psychological than anything else.

        You also asked my opinion on the relevance of the wealth effects, and my answer comes back to comments I made in the article about cash injections into the real economy versus the financial economy. When a bank loans money to, say, a developer to build a building, all or almost all of that money gets injected directly into the real economy (the builder uses it to pay people and buy things, so it increases spending). By comparison, when the central bank conjures money from nothing to create wealth effects, only a small portion of the extra wealth is used to boost spending. Studies show each dollar of extra wealth boosting spending by maybe a nickel or even less than that. In other words, the spending boost from wealth effects is an order of magnitude lower than the spending boost from injecting loan money directly into the real economy. So, my criterion for inclusion in M63 is that the creation of the particular type of money is initiated by a commercial bank or other deposit-taking and loan-making institution, not a central bank.

        • Andre


          Your answer is much appreciated. I have a feeling I’m really going to enjoy your book 🙂

  15. Pingback: What is Money? What is Inflation? | Icliks Incoming

  16. Pingback: An Inflation Indicator to Watch, Part 3 - Investing Matters

  17. Pingback: An Inflation Indicator To Watch: Part 3 – Earths Final Countdown

  18. Pingback: An Inflation Indicator to Watch, Part 3 –

  19. Pingback: An Inflation Indicator To Watch: Part 3 – Wall Street Karma

  20. Pingback: An Inflation Indicator To Watch: Part 3 | peoples trust toronto

  21. Pingback: An Inflation Indicator To Watch: Part 3 -

  22. Pingback: An Inflation Indicator To Watch: Part 3

  23. Pingback: An Inflation Indicator To Watch: Part 3 | Real Patriot News

  24. Pingback: An Inflation Indicator To Watch: Part 3 – ProTradingResearch

  25. Pingback: An Inflation Indicator To Watch: Part 3 – TradingCheatSheet

  26. Pingback: An Inflation Indicator To Watch: Part 3 |

  27. Pingback: Was The Fed Skillful Or Just Lucky? – ValueWalkPremium

Leave Comment

Your email address will not be published. Required fields are marked *