Deflating the inflation fear: an introduction to inflation in the modern monetary system

I explained in my last post the zero-sum relationship between net savings, public debt and the balance of trade, and that since government budget deficits do not need to be financed the only danger of deficit spending is high inflation. In this post I expand on that claim, introducing the widely misunderstood causes and effects of inflation and some of the dangerous misconceptions held by policymakers and reproduced in the media. This post is necessarily long, because to understand inflation requires an understanding of the monetary system; to understand where mainstream economics went wrong requires an understanding of the theory underlying it; to recognise good policy you must be able to recognise bad policy. To limit the length of what was becoming a very long post (which suddenly looks a lot shorter when transcribed from a word document into this blog theme!) I have not made direct reference to economic data sources. If you would like any claim in this article substantiated please ask me in the comments. The central banking model I present makes some minor departures from the system implemented in Australia by the RBA, more information on the Reserve Bank’s version of fiat money can be found on their website.


Almost all surviving schools of economic thought define inflation as an overall increase in prices, for varying interpretations of “overall”. Economists often use core inflation as a benchmark rate as it includes only goods and services with relatively stable prices. Volatile items including most consumer goods are excluded. One notable exception to the price level definition of inflation is used by economists of the Austrian school[1], which has maintained the classical tradition of defining inflation as an increase in the size of the money supply. Superficially these definitions may appear approximately equivalent: more money chasing the same quantity of goods will surely result in higher prices. In the long run and for carefully chosen interpretations of what counts as “money” this may prove true, but we are creatures of the short-run where factors of supply, demand and power are the main determinants of prices. In the short run the distinction is very important.

The precise relationship between the rate of money supply growth and the rate of inflation is hotly debated between competing economic schools. The dominant mainstream view today is called monetarism, which was developed most famously by economist Milton Friedman and influenced the economic policies of Jimmy Carter, Margaret Thatcher and Ronald Reagan, as well as Reagan’s appointed Chairman of the Federal Reserve Alan Greenspan and his successor Ben Bernanke, who continues to hold that position today. The monetarist position is that “inflation is always and everywhere a monetary phenomenon” (Friedman, 1970); it asserts that the powers of monetary policy – the central bank’s access to the printing press – are always sufficient to control price level growth.

Central Banking

The central bank is an organisation operated independently of the government which has been granted special legislative powers to control the money supply under the dual mandate of ensuring price stability and (in some cases) full employment. The privilege of being able to print money comes with the limitation of never really being able to spend it. Money printed by the central bank is only spent on loans[2], and only to the most creditworthy parties. The loans take the form of certificates called bonds, which are essentially fixed-term loans to be repaid to whoever holds the certificate. Traditionally central banks have only purchased bonds issued by the government, though more recently many have begun to accept bonds issued by highly creditworthy corporations under strict repurchase agreements[3]. As creditor for all currency in circulation the central bank is almost guaranteed to be profitable, though after all operational expenses have been paid the profits are usually paid back to the government to avoid the interest on the public debt rising forever.

Each of the major commercial banks has a reserve account at the central bank in which it must keep a portion of every deposit. These accounts pay zero or very low interest rates, so banks try to keep these the balance in these accounts as close to the regulatory minimum as they can. The government also keeps its accounts there, and it is through a trading platform at the central bank that government bonds are auctioned. The central bank also acts as a lender of last resort, buying bonds at the going rate plus a small penalty, if there are no other buyers for the debt. A commercial bank may borrow from the central bank overnight to meet reserve requirements if it is unable to borrow from other banks. Most reserve imbalances simply reflect transactions between customers at different banks, so the banks are usually able to meet their reserve requirements and offload excess reserves by lending amongst themselves. Banks may also buy government bonds with any excess reserves they might have, and the interest rate they charge on interbank lending is usually competed down to just above the interest rate they could receive by buying government bonds instead.

By participating in this market with unlimited buying power the central bank is able to exert influence over interest rates. Once it has decided on the base interest rate (called the cash rate in Australia) it announces the decision and then buys or sells government bonds as necessary to set the going rate. If it wants the rate to fall, it buys up any offer of bonds at a higher rate until only lower-yielding (i.e.: more expensive) bonds remain on the market. If it wants the rate to rise, it sells its holdings of government bonds until they are so prevalent that sellers must offer bonds at the desired higher yield (i.e.: lower cost) to successfully sell them. The interbank lending rate soon follows, since no bank will lend to another for a lower rate than it could earn buying government bonds yet any bank will accept even slightly more than that. Eventually the effect trickles down to retail interest rates, since being able to borrow more cheaply means that a bank can also afford to lend more cheaply.

By influencing the cost of borrowing money and the return from lending it monetary policy is able to encourage or discourage investment projects on the cusp of being profitable. If loose monetary policy over a period reduces interest rates by 2%, the revenue required for a million dollar business project to break even falls by $20,000. If the project goes ahead it generates demand for goods, services and labour, bidding the cost of these resources upwards through competition, hence creating an inflationary pressure. It is through this channel of interest rates and investment[4] that monetary policy attempts to control inflation.

Nothing in this article so far should be controversial. The model of central banking I have described is simplified but basically factual and resembles the textbook model taught in schools. Everything that follows is contested, though the textbook model which I will continue to describe is held by most of the economics profession (and many policymakers who studied economics) to be, as with the above description of central banking, simplified but basically factual.

Monetarism and the Mainstream Model

The fractional reserve banking system in the way it is usually taught expands the money supply through the following process:

  1. A customer deposits cash into her bank account.
  2. The bank keeps a percentage in reserves to facilitate withdrawals and satisfy regulations.
  3. The remaining amount (typically 90%) is loaned to another customer.
  4. The loaned cash ends up in other bank accounts and the process is repeated.
  5. After several iterations the lending capacity of the original deposit is exhausted.

At the end of the process the summed balances of all deposit accounts is much higher than the original deposit; the ratio between the larger amount (sometimes called ‘broad money’) and the original deposit (sometimes called ‘base money’) is called the money multiplier. Every new dollar of base money that enters the banking system is subject to this process of multiplication. The central bank encourages investment by buying up government bonds held by banks, which puts more reserves in bank vaults and enables them to make more loans before they reach the minimum reserve ratio. This is essentially the rationale behind conventional monetary policy: the key to stimulating economic activity is to “get banks lending again”.

If monetary policy has pushed interest rates to zero and still not achieved the desired level of investment – as it had in Japan in the mid-1990s and the US since 2008 – it must alter its approach. From the monetarist school of thought comes quantitative easing, wherein the central bank purchases a wider range of financial assets (debt) from banks and decides on a quantity of money to inject into bank reserves rather than a target interest rate. Since banks do not like to sit on idle reserves it is expected that this move will encourage them to extend more loans to investors and the general public.

Quantitative easing is not working. Neither Japan nor the US has yet been able to kick-start production, despite increasingly large injections of money by quantitative easing. Asset holdings at the US Federal Reserve have increased more than 280% since the quantitative easing programme began, rising from $700 billion to over $2 trillion in three years. Real GDP growth meanwhile has hovered between zero and one per cent since mid-2009 after spending most of the 18 months prior in decline.

The Truth about Bank Lending

The key insight to the failure of monetary policy to get the economy moving again comes from an off-mainstream school of thought called Modern Monetary Theory[5] (MMT), which emphasises accounting and certain operational realities of the banking system that have been ignored or forgotten in the mainstream. The reality MMT reminds us of is that bank lending is never constrained by reserves – at the end of the day a bank can always borrow from other banks or the central bank to meet its reserve requirements. If a loan is expected to be profitable the bank will make it, safe in the knowledge it can always find more reserves later – if not from the banks that now have excess reserves after the loaned funds have been disbursed, then from the central bank. This is a reversal of the traditional model, where deposits create loans until the reserve limit is met. In reality, loans create deposits until the demand for loans is met.

Quantitative easing is a solution to a problem that never existed. Trying to “get banks lending again” by making credit cheaper is like trying to get restaurants cooking again by giving them good deals on ingredients – if people aren’t hungry, it’s not going to work. The lack of hunger for loans in the US can be observed through the relationship between interest rates and inflation. The difference between them (interest minus inflation) is called the real interest rate and it represents the financial reward for saving. When the real interest rate is positive, money is in such high demand that investors must compete for the privilege of borrowing it from savers and willingly agree to pay more back in real terms than they borrowed. When the real interest rate is negative (as it is in the US), money is in such low demand that savers must compete for the privilege of lending it to investors and willingly agree to receive less back in real terms than they lent – any interest is better than full exposure to the erosive force of inflation.

The greatest failure of monetarism is that it sees aggregate demand (and hence inflation) as purely a function of the supply of money. Based on the same faulty model of bank lending, monetarists further insist that governments do not attempt to manage demand with fiscal policy. When the government wants to run a deficit it raises the cash by selling bonds on the open market, i.e.: by borrowing from banks. The monetarist position is that this is not expansionary because it merely ‘crowds out’ private investment: the government competition for funds pushes up the price of credit, raising interest rates and discouraging private borrowing. In reality two things happen when the government spends (the order is not important): the treasury issues bonds to the value of the spending it wishes to make, and that balance is credited at the central bank to the reserve accounts of the banks at which the recipients of that spending hold accounts. The banks find themselves with excess reserves that they cannot get rid of by lending amongst themselves, so instead they will attempt to buy government bonds, bidding the interest rate down in the process. This is why deficits do not need to be ‘financed’ – the injection of new reserves into the banking system generates a corresponding demand for government debt automatically.

Though monetarists usually deny that fiscal policy can be inflationary due to the crowding out effect, another perspective from the neoclassical school that monetarism belongs to is that fiscal policy will be excessively inflationary due to the money multiplier effect. Like other predictions that assume banks are waiting on more deposits with which to create new loans, it is bunk; like other mainstream economic theory there is a kernel of truth in the outcome, if not the reasoning that arrived there. There are three channels through which fiscal policy can be inflationary:

  1. Demand directly created through spending on public infrastructure projects.
  2. Demand created by recipients of fiscal transfers (e.g.: welfare recipients) to the extent that this income is spent, not saved.
  3. New income streams (interest payments on government debt) boost net private wealth, inflationary to a similar extent as above.

To understand the last point recall the relationship between net savings and government spending explained in my last post. Debt service payments are merely another form of government spending, contributions to net savings. This is an important nuance in the more comprehensive model used in this post: although monetary policy only creates credit and debt in equal amounts, the means for conducting it also create and destroy income streams in the form of interest payments, which affect fiscal outcomes. The same logic can be used to show an inherent limitation in the ability of monetary policy to create inflation without fiscal cooperation in the form of increased government spending: monetary policy may encourage business investment funded by credit money, but the profits will be taxed with real money, pulling net savings from the private sector and reducing net income. Fiscal policy must work with expansionary monetary policy to prevent this deflationary force from counteracting investment gains.

The Only Thing to Fear

Before going any further we need to be sure what kind of outcome we actually want. Living in a world where we are conditioned to be consumers it is easy to forget that we are merchants of our own labour. All prices are subject to inflation, including our wages: the price of our time. A common misconception is that inflation makes things more expensive. In fact there is no relative effect on the cost of goods in terms of other goods, except perhaps in the short-term before wages can adjust to a shock in some other part of the economy. There are some costs to particularly high rates of inflation: “menu costs” of keeping up with inflation, analogous to restaurants frequently reprinting menus with new prices, and “shoe-leather costs” of managing accounts, analogous to a store manager walking to and from the bank several times a day to deposit cash. People with lots of debt actually benefit from high inflation, because it reduces the real value of their loan and makes it easier to pay off. After the period of hyperinflation in Germany, and when inflation peaked at 25% in the UK in the 1970s, many homeowners found themselves with mortgage burdens much relieved or even paid off completely.  In this way high inflation can be viewed as a wealth transfer from (mostly rich) creditors to debtors[6]. At the lower end (below 5%) inflation serves an essential purpose for money: it encourages people to either spend their earnings or lend it to someone else who will – anyone who tries to keep their money to themselves finds its value eroded away over time.

There is a human factor to inflation which is far more powerful than any combination of monetary and fiscal policy. Expectations – the way confidence and uncertainty manifest in market behaviour – are the entry point for “real economics” into what has so far been largely an accounting model of the macroeconomy. They are the result of the tendency for people to try and stay ahead of the market and all too often their prophecy is self-fulfilling. When the inflation rate is high, expectations of higher costs in the future may result in merchants trying to stay ahead by raising prices a little more today. Since they have all done it the prediction is correct, and the cycle continues, potentially leading to hyperinflation if the supply of savings is sufficiently large, as it is when the public debt is very large relative to GDP. During the height of hyperinflation in Germany (so the story goes) you could leave a wheelbarrow full of money out on the street at night and come back to find someone had dumped out the cash and stolen the wheelbarrow. Hyperinflation is difficult to control, but it is only a danger to countries running very high deficits – usually, when a government fighting a losing war on home turf decides that high inflation is preferable to defeat.

Though inflation must be high and out of control to be problematic, the danger is asymmetrical. Even a miniscule amount of deflation, if it triggers deflationary expectations, will be much worse: with prices falling, any nonessential spending is likely to be put off, since it is expected to be cheaper in the future. Employers who expect reduced demand in the future are less likely to hire new workers and more likely to lay off existing staff, since it is virtually impossible, politically, to reduce workers’ nominal wages even if deflation keeps the real wage growing. Borrowing for investment becomes impractical too, as deflation adds to the real interest rate of the loan and weak demand makes it more difficult to earn sufficient revenue.

It is through this channel of expectations that monetarism brings substance to the claim that inflation can always be controlled with monetary policy. The reasoning is complex and internally (mostly) logically consistent. It works something like this:

  1. Assume that the economy begins in equilibrium, i.e.: all private demand for savings has been met.
  2.  Monetary policy increases the size of the money supply, creating new deposits.
  3. The new money being held is surplus to desired savings; if not saved it must be spent.
  4. Expectations of future demand cause business to respond to loose monetary policy with expansion and investment.

Note here that the use of expectations negates in step four effectively negates the problems we identified earlier in the efficacy of monetary policy in step two. Though unappealing at face value to accept a theory which appears to require a kind of circular logic, begging the question of why loose monetary policy is expected to be expansionary in the first place, the point is moot; the most damning problem is in the very first assumption – in fact, the first assumption of all neoclassical economics – that the macroeconomy is ever in a state of equilibrium. A thorough breakdown of the concept of equilibrium is beyond the scope of this post, but the essence of the idea – the state where forces of supply and demand are balanced across the whole economy – even at a casual glance seems antithetical to the need for monetary intervention at all. Though general equilibrium has been shown to be unstable or infeasible in both academic and professional publications, proponents of neoclassical and monetarist methods have clung to it, asserting that a model should not be judged by the realism of its assumptions but by its experimental performance. Evaluating the performance of the monetarist approach in the current economic climate is left as an exercise for the reader.

This article barely scratches the surface of the complex phenomenon of inflation – but if you have read this far and followed along then you are already at an advantage compared to most political and economic commentators, who view government spending through the same lens as the spending of an individual. They see a balanced budget as the paragon of fiscal responsibility and when the powers that be fail to restart the faltering economies in the US, the UK and Europe they write that no one knows what is going on. The reality is that the answers are out there for those willing to listen. Economist Robert P. Murphy rejected the insights of Modern Monetary Theory on its basis in accounting, stating that “it’s bad economics to confuse accounting identities with behavioural laws […] economics is not accounting.”[7] And yet in a study that found only 13 people who had accurately reasoned and predicted the financial crisis of 2008, the common theme in all their work was an accounting emphasis[8]. I hope that by sharing some of their insights I can help demystify the mystic depths of macroeconomics for others, and with enough time and luck perhaps shift the focus of political debate from the imaginary economic problems prevalent in contemporary discourse to real ones.

[1] Referring to a collection of economic theories first proposed in Austria. Most Austrian economists today are American.
[2] For the purposes of monetary policy. The central bank may purchase currency and other assets outright, e.g.: when it chooses to participate in foreign exchange to set the exchange rate.
[3] Also known as repo contracts, these are agreements that the seller of a bond (not necessarily the issuer) will buy it back after a certain period.
[4] Another important variable here is unemployment, which will be the topic of a future post.
[5] Modern Monetary Theory is one incarnation of the branch of macroeconomics known as chartalism (from the Greek ‘charta’, meaning ‘token’). It is the study of ‘token currencies’ as distinct from currencies made from or backed by commodities.
[6] The reverse situation of deflation, rising real value of debt can trigger mass default. This could also be viewed as a wealth transfer from creditors to debtors, albeit with a much more damaging aftermath.
[7] See: Kuehn, Daniel (2011). “Murphy on the MMTers”.
[8] See: Bezemer, Dirk J (2009). ““No One Saw This Coming”: Understanding Financial Crisis Through Accounting Models

3 comments to Deflating the inflation fear: an introduction to inflation in the modern monetary system

  • Porridge

    “In reality, loans create deposits until the demand for loans is met.”

    I can visualise how the reverse of this works, ie: the traditional model, but I’m having real difficulty wrapping my head around the line above. Are these deposits loan repayments, or savings, or some other thing?

  • [law]

    Imagine you’re a bank manager and your bank is already at the lower limit of its reserves – you’ve lent out as much of your customers’ savings as you can without breaking the regulations. An important corporate client comes to you one afternoon and asks for a loan which you stand to make a lot of money on. Your options are: turn the important client away, or give him the loan anyway knowing that you can always borrow cash from another bank if you need it. You know that you’ll be able to, since if he takes the money to another bank that bank will have a whole bunch of excess reserves to get rid of, and the central bank is always there as a last resort too. Of course you make the loan.

    Now imagine that every bank is doing this. At any given time, there are no loans waiting on reserves because there’s no risk of not being able to find reserves. That’s what I mean by ‘loans create deposits until the demand for loans is met’ – banks just keep loaning until there’s no more demand for loans. The loaned money gets paid out to people, always ending up in someone’s savings account ready to be lent out again.

  • […] paid on government bonds is reduced by a reduction in the amount of government debt. MMT provides a model for understanding why issuing government debt should in fact quickly cause the interest rate to […]

Leave a Reply




You can use these HTML tags

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>