Recent Posts

Thursday, June 4, 2015

Understanding Government Finance - Released

The eReport Understanding Government Finance has been released, and is available at online retail stores in ebook format. The current list of retailers includes Amazon, Kobo, the Apple Store, Nook, Scribd, and the Tolino network. The widget above gives an online preview (from Amazon).


The government budget is not like a household budget. This report introduces the financial operations used by a central government with a free-floating currency, and explains how they differ from that of a household or corporation. The focus is on the types of constraints such a government faces, and what can cause such a government to default. This report also acts as an introduction to some of the concepts used by Modern Monetary Theory, a school of thought within economics. Modern Monetary Theory emphasises the real limits of government action, as opposed to purely theoretical views about fiscal policy.

  • Book length: around 34,000 words. Paperback is 122 pages (103 excluding front matter).
  • 18 Figures and charts.

Paperback Edition - ISBN 978-0-9947480-5-8

The paperback edition was released July 4th, 2016. The book will be available at online booksellers, and via special order from bookstores. It will take 5-8 weeks to appear in some channels; it will appear first on Amazon sites.

There have been small differences from the ebook editions as a result of the format change, including the addition of figure numbers and an index.

Once the paperback and Kindle edition of the book are linked (within a few days), it should be possible to buy the paperback and receive the Kindle edition for free (the "book matching program").

EPUB Format - ISBN 978-0-9947480-0-3

There are desktop and tablet apps to read the book, which are available at the retailer's sites. Also, there is Adobe Digital Editions, which allows you to read epubs, and manage them across a several portable devices.

Kindle (Amazon) Format - ISBN 978-0-9947480-1-0

Tablets have a free Kindle app, you can read the book via a web browser ("on the cloud"), and there is also is a free app to read on desktop computers.

Amazon Stores:

Table Of Contents

Chapter 1  Overview
1.1  Introduction
1.2  About this Report
Chapter 2  Understanding Money
2.1  Introduction
2.2  Money Is a Unit of Account
2.3  Chartalism
2.4  Money and Bonds
2.5  Concluding Remarks
Chapter 3  Introduction to Government Finances
3.1  Introduction
3.2  Constraints within Economic Models
3.3  What Is a Central Bank?
3.4  Simplified Framework of Government Finance
3.5  Consolidation of the Treasury and the Central Bank
3.6  Concluding Remarks
Chapter 4  Government Financial Operations
4.1  Introduction
4.2  The Limited Role of Currency
4.3  Transactions Not Involving Currency
4.4  Can the Treasury Run Out of Money?
4.5  Concluding Remarks
Chapter 5  Extensions–Reserves, Government Lending
5.1  Introduction
5.2  Bank Reserves
5.3  Government Lending
5.4  Aside: Quantitative Easing
5.5  Concluding Remarks
Chapter 6  Bonds and Interest Rates
6.1  Introduction
6.2  Bond Yields and the Debt-to-GDP Ratio
6.3  Marginal Analysis and Bond Yields
6.4  Central Bank Control of Short Rates
6.5  Rate Expectations and the Term Premium
6.6  Rollover Risk
6.7  Why Issue Bonds?
6.8  Concluding Remarks
Chapter 7  Domestic Constraints on Deficits
7.1  Introduction
7.2  Functional Finance
7.3  Mainstream Analysis of Government Financing
7.4  The Governmental Budget Constraint
7.5  Paying the Debt Back?
7.6  Complications with the Budget Constraint
7.7  Can We Model Fiscal Constraints?
7.8  Concluding Remarks
Chapter 8  Outside Constraints on Governments
8.1  Introduction
8.2  Borrowing in Foreign Currency
8.3  Gold Parities and Currency Pegs
8.4  One-Sided Currency Pegs
8.5  External Constraints without a Currency Peg
8.6  Sub-Sovereigns
8.7  Concluding Remarks
Chapter 9  Conclusions
A.1  Appendix: Government Bonds and Bills
A1.1  Introduction
A1.2  Treasury Bills
A1.3  Forwards
A1.4  Bonds
A.2  Appendix: Overdraft Economies
A.3  Appendix: Rate Expectations and the Term Premium
Data Sources
References and Further Reading
About the Author

Final Comments

I look forward to receiving your support and feedback. Comments here are highly welcome, as well as reviews on retail web sites.

(c) Brian Romanchuk 2015


  1. Will it get released in PDF format? I unfortunately do not have an e-reader. Thank you.

    1. Hello,

      The short answer - yes it will, but I need to find a distribution channel, as well as convert the document to PDF format.

      The problem with PDF (for me) is that the page layout is fixed. I need to do the same layout work that is required to create the paperback version. I will tackle that problem "soon".

      But I would point out that there are free apps for tablets (e.g. iPad), as well as for desktop computers. Obviously, PDF is more convenient for desktops, but you will have to purchase it from a less familiar web site. (I will not directly fulfil orders as that creates a taxation nightmare for me.)

  2. Thanks for the quick reply. I think I'll give the Kindle version a try, and try to convert it myself to PDF (if I figure out how).

    1. Calibre is a free ebook management and conversion system. I do not know whether it can handle the Kindle version, but I have used it to convert the EPUB version to a PDF. (The downside is that the page size is small, and you end up with a lot of pages.)

      You can also read the Kindle version in a browser ("on the cloud").

      Unfortunately, the EPUB release date was set to Monday (June 8). I will check with the distributor to see whether that can be moved up.

  3. I signed up for Kobo, which gave a $5 credit, so I obtained the eReport as my first purchase without having to enter credit card information. Also, although I don't like most third-party digitial media managers (just a list of files and tools for access would be fine for me) the Kobo reader for Windows is adequate.

    My first comment/question is with regard to the imbalance of safe money market instruments described in section 2.2 Money is a Unit of Account page 5/12. Although it appears on the surface that money market mutual funds (MMMF) are alternatives to bank deposits, in reality, when a depositor moves funds to an MMMF the first order impact is the MMMF now holds deposits in a bank. I think the next step is the aggregate bank engages in repurchase agreement borrowing from the MMMFs, which benefits both banks and MMMFs in their credit operations, so the deposits appear to disappear in aggregate. I know this type of analysis may be beyond the intended scope of the eReport, however, it does raise a question in my mind as to how the imbalance of supply and demand occurs in money markets. The key is understanding that if the aggregate bank wants to expand it would simply borrow from money market managers, and cancel deposits, when deposits are moved to MMMF funds, forming a daisy-chain of liabilities back to the aggregate bank balance sheet. The aggregate bank is in crisis due to roll-over risk of the daisy chain in repo and euro dollar liabilities.

    1. Thanks for your support. And that's good to hear about the sign up bonus. Adobe Digital Editions is useful if you are managing an e-reader (I have a Kobo, which is popular in Canada), or files with DRM. (My ebook does not have DRM.) Calibre also has a very good Epub viewer.

      I tried to avoid being dragged into banking system controversies. If I had been aiming for a full length book I would have added in a discussion of banking, but the report ended up being at about the upper limit of the word count I am targeting. I will probably create a mini-report which is a banking system primer, which I will either price very cheaply (.99) or free. (Banking system controversies come up a lot.)

      As for your question, the imbalance is between the desired portfolios of investors, and the profile of desired borrowings. There is a large demand for safe "money-like" instruments, but very few high quality borrowers want to have large amounts of short-term debt. (If they did, they may no longer be quality borrowers.)

      If there was only the banking system, this maturity mismatch is managed on bank balance sheets. But when we add in the non-bank financial sector, deposits now migrate to "shadow banking" (including vanilla MMFs). The non-bank system can expand its balance sheet in a similar fashion to the formal banking sector; you could treat a migration of a deposit to a MMF as being equivalent to an inter-bank transfer.

      The imbalance of how people want to hold "deposits" (including MMF holdings) versus borrowing desires means that the shadow banking system has to engage in maturity transformation. The banks have the central bank to backstop when this breaks down, but the MMF's were not expected to be backstopped. (During the crisis, they were also bailed out.)

    2. Thanks for the feedback. See Transactions 1-4 of this Fed paper from 1980 to illustrate exactly what I mean by a daisy-chain of Eurobank liabilities back to the aggregate bank balance sheet. When a depositor moves funds to a Eurobank (MMMF) the bank cancels a deposit account and issues a wholesale bank liability to the Eurobank (MMMF). So these so-called deposit alternatives do not mean a loss of bank liabilities simply a change in the mix of bank liabilities. This daisy-chain relates to the issue of how "shadow bank" activity transmits financial fragility to the banking system forcing the central bank to act as lender of last resort to MMMFs, Eurobanks, banks, etc.

      I agree with your approach to focus on the nature of government finance operations in a simplified yet accurate model and look forward to reading the remainder of the eReport.

  4. Brian, below is a terse attempt to alert you to a point of confusion while reading section 4.

    Section 3.5 pgs. 16-17 show a Simplified Framework for consolidated govt & central bank balance sheet:

    FixedAssets - GovtBonds - Currency - Capital = 0

    Section 4.1 page 1 says transactions between consolidated government and private sector must be settled in government bonds, however, this introduces confusion because nonbanks pay for bonds and taxes via the surrender of transaction accounts (bank liabilities) to the Treasury which tend to increase the FixedAssets, and the Treasury spends funds back to nonbanks as transaction accounts. This means the Treasury can tax and borrow before it spends and the private sector does not have to surrender old bonds to purchase new bonds as shown in section 4.3 page 10.

    1. In the Simplified Framework, the central bank (and Treasury) do not lend to the private sector. And since I also eliminate reserves, that leaves the only way for the private sector to transfer "money" to the government is via selling bonds or currency (notes and coins).

      Non-banks are assumed to use banks as intermediaries. The banks would be forced to find the settlement balances, and then the non-bank settles its accounts with the bank.

      If the non-bank entity pays for something by cheque (check), this only introduces a delay in the settlement process. If the cheque is not honoured by the bank ("bounces"), the government will no longer consider the non-bank's account to be settled, and will ask for payment again. I will consider adding a discussion of the issue of "cheque float". The issue is that if I do so, I need to start discussing how the private banking system operates.

    2. If new bonds are always paid by old bonds how do any bonds get into circulation in the first instance?

      Historically gold reserve banking permits banks to create inside money and the treasury can tax, borrow, and spend to recycle inside money in the presence of absence of a central bank. Thus government bonds are net generated and there is no requirement to sell bonds to the government to purchase bonds from the government because inside money is generated via a parallel process and government finance recycles inside money.

    3. The spending operations draw bonds out from the central bank, or they could trade in currency. Currency is not usually used, but it is possible.

      There is the question of what happens at "time zero" - how does the system start out? In the real world, modern countries started from gold-based systems that developed from the ruins of previous civilizations. If you somehow started a new country from scratch, you probably would be working in a quite unusual financial environment. About the only way this could be achieved is starting a colony on an island, with no external trade.

    4. I think Einstein said "Everything should be made as simple as possible yet no simpler." I use a slightly more complex simplified balance sheet for combined central bank and Treasury in the United States:

      NetWorth = NFA + FA* + TTL - HPM - GovtBonds - OtherLiabilities

      where the government holds nonfinancial assets (NFA) and financial assets as the sum of FA* and Treasury Tax & Loan TTL (United States term). The FA* include financial assets of central bank and treasury which are liabilities of banks and nonbanks, but FA* excludes TTL, which are bank liabilities used by Treasury to accept payment by nonbanks in cooperation with banks.

      High powered money HPM is controlled by the central bank and includes the sum of currency issued to banks and nonbanks as well as bank reserves on account with the central bank.

      Prior to 2008 the Treasury sells GovtBonds to banks and nonbanks and Fed does not accumulate such bonds on its balance sheet except to offset the level of currency in circulation with HPM and to provide banks with a small amount of required plus precautionary reserves.

      There is a vertical and horizontal chart shown by Bill Mitchell here:

      where the government seems to be able to hold HPM constant while increasing the float of GovtBonds in the "tin shed" via deficit spending. If taxes exceed spending over a period the government retires some GovtBonds from the "tin shed." When nonbanks want to hold more currency the central bank purchases an equal amount of GovtBonds from the "tin shed."

      I appreciate your efforts to simplify the system, however, it should be possible to discuss a simplified model in which the government debt can increase via deficit spending without any large net change in levels of high powered money or bank balances, which is how the system operates to support both a government deficit and independent central bank control over interbank interest rates.

    5. If you look at the system I describe, deficit spending raises government debt levels without affecting high powered money. It is done in two operations - spending, and then bond auctions. This exactly the same as the standard MMT description, except I have eliminated bank reserves. In the standard MMT description, spending increases reserves, which are then drained. This does not match up to the a system such as Canada, where reserves do not exist.

    6. I still find your description confusing in section 4. In my interpretation if reserves (bank settlement funds) do not exist in a system then government spending or central bank lending cannot generate the positive settlement balances at the central bank which are necessary to purchase new bonds without having to surrender old bonds. So reserves must exist to permit the increase of the float of government bonds unless markets accept government bonds directly in payment (rather than money) in exchange for goods and services sold to government.

      William F. Hummel says the Treasury effectively pays for goods and services under deficit spending by issuing securities directly to markets. In the United States this also requires the Treasury to recycle preexisting inside money in the form of bank reserves and transaction accounts with an effort to not disturb the reserve levels which assists the Fed with control over the interbank interest rate in fed funds markets. Perhaps the Fed must inject daylight credit (reserves) to support the Treasury recycling operations but at the end of the day the money injected is gone and the Treasury float increases as the net result, so it does not appear as though government must spend before it taxes or borrows, it simply must coordinate these efforts closely to keep from draining banks of reserves by taxing/borrowing without matched spending.

    7. I am not describing the operations as they are used in the United States right now. I am describing a system in which bank reserves do not exist, such as in Canada.

      The complications that appear to concern you are discussed in Chapter 5, which add in reserves and lending from governments. Neither feature is necessary for a fiat monetary system to function. Banks can use old bonds (or more likely, bills) to manage their liquidity positions. Banks can run a non-zero settlement balance during the day, but it has to be settled to be zero at the end of the day.

    8. I do not know the details of the Canadian scheme. You are saying that banks and primary dealers need positive settlement funds to purchase government bonds. In the United States this means reserves EXIST and are provided by the government at least during the daylight credit hours. If there were zero overnight reserves it means the consolidated government drains the reserves to zero or forces some banks to overdraft with the central bank while other banks hold an equal and opposite positive balance.

      In the United States such a system would have a different diagram than the ones you show. The government would pay the private sector directly with government bonds in exchange for goods and services. More specifically, Treasury or Fed would debit an asset (+) and credit government bonds (+) to acquire nonfinancial or financial assets on the consolidated balance sheet. When banks want to purchase currency from Fed, they would have to pay Fed with old government bonds. If you are asserting that banks do not own or use reserves to clear payment (reserves do not exist) there is no reason for the private sector to sell old bonds to the government other than to obtain currency since there are no exchange settlement balance positions to manage if they don't exist!

    9. If you re-read what I wrote in Section 4.1, you will see that I said that settlement balances at the central bank can be either positive or negative during the day, but the position has to be settled to zero at the end of the day. The banking system "pays" for the bonds at auction by running an intraday negative balance, and then they have to transfer bonds and bills at the end of the day to get the balance back to zero. I note several times that I am discussing net transactions, and that the transfers of bonds/bills are a netted end-of-day settlement.

      In the United States pre-2008, pretty much the same thing happened, except that "excess reserves" replaces "settlement balances" in the above description. I point this out in Chapter 5, when I add the complexities associated with the American system.

      I will probably be publishing excerpts of the book over the coming weeks, including a description of the Simplified Framework. I think it will make more sense to continue the discussion then. Unless the other readers of the comments have read my book, they have no idea what this discussion refers to, since I have not published this content anywhere else.

    10. Great, I will refrain from further comment here. I just want to add here that since a central bank manages its balance sheet to provide reserves (positive interbank settlement funds) to banks at the target interest rate, if there are zero overnight reserves in the system, then the Simplified Framework the overnight central bank assets (government bonds) should be equal to the currency issued plus capital. If one is abstracting away from reserve assets and transaction account money as means of clearing payments then the only reason the private sector has to send old bonds back to government is to purchase currency or settle liabilities such as taxes and other accounts payable to government.


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

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

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