Friday, December 1, 2017


Comview 2017

As the monetarists used to like saying, monetary policy operates with “long and variable lags”. Which does much to explain why, for most of the time I’ve been an economic journalist, people have doubted its effectiveness. In 1989, when the Hawke government was struggling to slow a strong economy and booming commercial property market, with the cash rate hitting a peak of 18 per cent and mortgage rates a peak of 17 per cent, I remember a colleague writing it was clear that monetary policy had lost its power to dampen demand. That, of course, was just before the economy plunged into the severe recession of the early 1990s, its longest and deepest slump since the 1930s.

But the fear that tight monetary policy has lost its power to slow the economy is much rarer that the opposite fear that easy monetary policy has lost its power to speed the economy. Remember the old Keynesian jibe that cutting rates to stimulate demand is like “pushing on a string”? Until, of course, we’ve worked through the long and variable lags and discover the low rates have indeed boosted demand. But the fear that monetary policy has lost its power is much easier to credit in recent times.

Consider the facts. In November 2011, when the cash rate was at 4.5 per cent, the Reserve Bank decided it could stop worrying about an inflation surge and should cut rates to ensure continued trend growth. By December the following year, it had the rate down to 3 per cent. By August 2013 it was down to 2.5 per cent. It waited more than a year and a half to reduce the rate to 2 per cent by May 2015, then waited another year before making two more cuts to its present 1.5 per cent in August 2016. It’s maintained that record low for the 15 months since then. So the Reserve has cut the official interest rate by 3 percentage points on and off over the past six years, but the economy’s growth has been below trend for almost all of that time, leaving the present case for doubting monetary policy’s efficacy looking persuasive.

The big questions is why? We’ll get to that, but first I want to give you an update on the “framework” for monetary policy, the monetary policy “transmission mechanism”, including the “channels” through which monetary policy affects economic activity and inflation, and how we assess the “stance” of monetary policy using the “neutral” interest rate.

The monetary policy “framework”

Monetary policy - the manipulation of interest rates to influence the strength of demand - is conducted by the Reserve Bank independent of the elected government. It is the primary instrument by which the managers of the economy pursue internal balance - low inflation and low unemployment. Monetary policy is conducted in accordance with the inflation target: to hold the inflation rate between 2 and 3 per cent, on average, over time. The primary instrument of monetary policy is the overnight cash rate, which the Reserve controls via market operations.

The monetary policy transmission mechanism

The mechanism by which monetary policy affects economic activity (production and employment) and inflation can be divided into two stages: first, changes in the cash rate affect other interest rates in the economy and, second, changes in these in these interest rates affect economic activity and inflation.

Taking the case of an “easing” in policy, stage one involves the Reserve Bank deciding to lower the overnight cash rate (the market interest rate for overnight loans between financial institutions; also known as the “official” interest rate, or the “policy” interest rate). This causes financial markets to change their expectations about the future path of the cash rate and the structures of deposit and lending rates are quickly altered. The effect on short-term and variable interest rates happens much earlier than the effect on long-term interest rates.

Stage two involves households and firms responding to lower interest rates by increasing their demand for credit, reducing their saving and increasing their current demand for goods, services and assets (such as housing and shares). Other things equal, rising demand increases the prices of non-tradable goods and services. The rising demand tends to raise the cost of inputs, including labour, leading to higher inflation. According to estimates by the Reserve, lowering the cash rate by 100 basis points (1 percentage point) leads to real GDP being ½ to ¾ of a percentage point higher than it otherwise would be over the course of two years. Inflation typically rises by a bit less than ¼ of a percentage point per year over two to three years. It takes between one and two years for changes in the cash rate to have their maximum effect on economic activity and inflation.

The main “channels” through which interest rates affect activity and inflation

  1. The saving and investment channel (also known as the “inter-temporal substitution effect”). Whether you are a saver or a borrower, interest rates are the opportunity cost of choosing to spend now rather than later. So lowering interest rates encourages households and businesses to reduce their saving or increase their borrowing so as to increase their spending on consumption or investment goods. Inter-temporal means “between time periods”, so reducing interest rates encourages people to bring their consumption and investment spending forward in time, whereas raising rates encourages them to push their spending off into the future. This (very neoclassical) channel is the main one referred to in textbooks and used in economic modelling of the economy. Note, however, the Reserve’s observation that “there is mixed evidence as to whether a strong relationship between lower interest rates and higher consumption growth actually exists”.

  2. The cash flow channel. Lower interest rates influence the spending decisions of households and businesses by reducing the amount of interest they pay on debt and the interest income they receive on deposits. This affects the disposable income (or cash flow) they have available to spend. Clearly, lower rates have opposite effects on borrowers (those with more variable-rate debt than deposits) than lenders (those with more variable-rate deposits than debt), with borrowers having more cash to spend and lenders having less. Even so, the effect on borrowers outweighs the effect on lenders, so that a fall in rates leads to increased spending. This is because the household sector is a net debtor, with the average borrower household holding two or three times as much debt as the average lender household holds in interest-earning deposits. But also because the spending of borrowers is more sensitive to changes in cash flows than the spending of lenders. The Reserve estimates that lowering the cash rate by 100 basis points increases total household disposable income by about 0.9 per cent, which then increases household spending by about 0.2 per cent.

  3. The wealth channel. A reduction in interest rates stimulates demand for assets such as shares and housing, raising their prices. This is because the lower rates increase the present discounted value of the assets’ future stream of income. Higher asset prices increase the wealth of households and businesses, which may lead them to increase their spending because they feel wealthier.

  4. The exchange rate channel. When interest rates fall (relative to those on offer in other countries) this attracts less net inflow of foreign capital, which lowers our exchange rate. This improves the international price competitiveness of our export industries, as well as making it easier for our import-competing industries to make sales in the domestic market. So lower interest rates should increase the growth in our production (GDP). It also adds to inflation directly by increasing the price of imports. But Reserve estimates suggest the effect of interest rates on the exchange rate is relatively small, with an unexpected 25 basis point decrease in the cash rate estimated to lead to a ¼ to ½ per cent depreciation in the exchange rate. Other estimates suggest a 10 per cent depreciation increases the volume of exports by 3 per cent, while reducing the volume of imports by 4 per cent, within two years.

Assessing the stance of monetary policy using the neutral interest rate

While it’s easy to see that a cut in interest rates represents a move in a less restrictive, more expansionary (“accommodating”) direction, and a rise in rates represents a move in a more restrictive, more contractionary direction, this doesn’t tell us anything about whether the present level of interest rates is expansionary or contractionary – known as the “stance” of policy. We need a benchmark against which to determine whether the present level of rates is expansionary or contractionary. The benchmark we use is the “neutral” interest rate, the rate that’s neither expansionary nor contractionary. Comparing the cash rate (policy rate) with the neutral rate also shows us the degree to which the policy stance is expansionary or contractionary.

The neutral interest rate is the real policy interest rate required to bring about full employment and stable inflation (practical “price stability”) over the medium term. But what factors influence the level of the natural rate? Assuming a closed economy, all investment must be funded by domestic saving. The neutral interest rate equilibrates saving and investment at the level of income that is consistent with full employment and stable inflation. This means developments that increase saving will tend to lower the neutral rate, while developments that increase investment will tend to raise the neutral rate. This, in turn, means the neutral rate is influenced by three main factors: 1) the economy’s “potential” growth rate, 2) the “risk appetite” of the economy’s firms and households, and, since we actually live in an open economy, 3) global interest rates.

Our potential growth rate is determined by the three Ps – population, participation and productivity, it being a supply side concept. Our rate of population growth is high relative to other developed countries’, as is our rate of productivity improvement, so we have a relatively strong incentive for firms to invest, implying a higher neutral rate than the others. Like the others, our participation rate is reduced by the ageing of the population (retirement of the baby boomer bulge), but we have more scope for increased participation by older women. 2) When aversion to risk increases, firms become less willing to make long-term investments with uncertain return. This reduces the demand for borrowed funds, while increasing households desire to save, and so tends to lower the neutral rate. 3) Global interest rates have a big effect on our neutral rate because we are so open, but also such a small part of the global economy. Strong Australian demand for investment funds will attract capital inflow (of foreigners’ savings), pushing up our exchange rate, but also limiting the extent to which our neutral rate exceeds the world rate.

Because the neutral interest rate is “not directly observable” (just as it’s close relatives, the NAIRU and the potential growth rate, aren’t either) they have to be estimated by economists, meaning that different economists will reach different estimates. According to the Reserve’s estimates, our neutral rate was fairly stable at about 3½ per cent from the early 1990s until 2007 at the start of the global financial crisis. Since then it has declined steadily and is now about 1 per cent. Remember, these figures are real. Add an expected inflation rate of 2.5 per cent (mid-point of the inflation target range) and you get a nominal neutral rate of 3.5 per cent.

The Reserve says most of the decline of about 250 basis points since 2007 is explained by a decline in our potential growth rate (about 50 basis points) and an increase in risk aversion (100 points). Similar factors have been at work in the major developed countries, meaning their neutral rates have fallen to a roughly similar extent.

Why monetary policy is less effective in recent years

The continuing below-trend economic growth despite a major easing in monetary policy, and plenty of time for it to work its way through the economy, suggests monetary policy easing no longer has as much effect as it used to in stimulating demand. Similar conclusions drawn in the major economies may be explained by their need to resort to the less-effective quantitative easing once official interest rates had been cut to zero. But that doesn’t apply to Australia, and there is no reason to suppose monetary policy has become less effective simply because interest rates here are a lot lower (closer to zero) than they used to be.

In his last speech before retiring in 2016, the former Reserve Bank governor, Glenn Stevens, said he’d long held the view that monetary policy’s main effect on demand was via households, rather than businesses. This was because businesses’ decisions about investment were influenced more by their assessment of the outlook for growth and profits than by the cost of capital – interest rates. So the main channel through which expansionary monetary policy works is to use lower interest rates to encourage households to borrow and spend more. Stevens then argued that this hadn’t been as effective in recent years because our very high level of household debt (most of which is for housing) was making people reluctant to borrow a lot more.

It seems clear the new governor, Philip Lowe, agrees with this assessment. He has made the important point that monetary policy’s reduced effectiveness is likely to be asymmetrical: if households’ high debt stops cuts in interest rates from encouraging much additional demand, this should mean that increases in interest rates were a lot more effective in discouraging demand (because households’ high levels of debt mean a rise in rates causes a bigger hit to their cash flow).

There is little doubt that the long period of unusually low mortgage interest rates has done much to encourage increased borrowing for housing, particularly in Sydney and Melbourne, making already high levels of household debt even higher. House prices have risen at huge and worrying rates, with competition from housing investment buyers making it a lot harder for young people to afford their first home. In other state capitals, however – notably, Perth – house prices have been weak. This is a reminder of one longstanding drawback in using monetary policy to control demand: you can have only one, uniform interest rate for the whole economy, even though demand is too strong in some states and too weak in others.

There is continuing speculation in markets and the media on whether the Reserve will cut rates further – to get demand growing stronger and inflation back up into the target range – or whether it will start raising rates to stop the rapid rise in house prices and Sydney and Melbourne. My guess is the Reserve wouldn’t mind being able at do both at the same time. Since this is impossible, it is pleased to have help from “macro-prudential” measures taken by the bank regulator, the Australian Prudential Regulation Authority, APRA, in tightening its direct controls over banks’ lending for investor housing.

How fiscal policy can help

Dr Lowe, has stepped up pressure on the Turnbull government (echoed by the IMF and OECD) for fiscal policy to give more assistance to monetary policy in encouraging demand. The government has been preoccupied with achieving fiscal policy’s primary goal of “fiscal sustainably” (ensuring the level of government debt doesn’t get too high) by attempting to get the budget back to surplus - though with little success because of the weak growth in tax collections.

Dr Lowe has argued that the government should draw a clearer distinction between its spending for capital (infrastructure investment) and its spending for recurrent (day-to-day) purposes. It should focus on getting only the recurrent or “operating” balance back to surplus, which would leave it free to give more support to demand, as well as do more to improve productivity, by continuing to borrow for worthwhile infrastructure projects. In this year’s budget the government responded to this pressure, giving more prominence to the net operating balance – the NOB - and by initiating two big infrastructure projects, the second Sydney airport and the Melbourne to Brisbane inland freight railway, with more capital city road and rail projects to come.


The Transmission of Monetary Policy: How Does It Work? Tim Atkins and Gianni La Cava, Reserve Bank Bulletin, September quarter, 2017

The Neutral Interest Rate, Rachael McCririck and Daniel Rees, Reserve Bank Bulletin, September quarter, 2017