Stuck in lowflation

Australia is still stuck in lowflation. Dwelling approvals continue to fall indicating ongoing weakness for construction although there was more positive news for the housing sector with the first monthly rise in the CoreLogic combined capitals index since late 2017. The US Fed cut rates by 25bp in the first downward move since 2008, but neither markets nor President Trump were impressed. This week’s readings cover possible explanations for low wage growth in Australia, the evolution of globalisation, London’s struggles with dirty money, and Latin America’s economic travails.

The annual rate of consumer price inflation was 1.6 per cent in June, high enough to convince markets that the RBA is unlikely to cut the cash rate next week but low enough to leave the economy stuck in lowflation.

Dwelling approvals fell in June, signalling more weakness in dwelling investment ahead and continued tough times for the construction sector.

Credit to the private sector rose 0.1 per cent over the month in June, with a small increase in credit to housing but there were monthly falls in credit to the business and personal sectors.

Dwelling values rose by 0.1 per cent in July, ending a run of 20 consecutive months of falling values.

The US Fed delivered its first rate cut since 2008, cutting rates by 25bp on 31 July. Fed Chair Jerome Powell also announced that the central bank would end the run-off of its asset portfolio (‘quantitative tightening’) two months earlier than planned.

The US-China trade war intensified after President Trump said the US would impose a 10 per cent tariff on $300 billion of additional Chinese goods.

The British pound has been sliding as markets price an increased risk of a no-deal Brexit.

(We’ll probably look at the Australian retail sales result, due on Friday, in next week’s piece.)

What I’ve been following in Australia . . .

What happened:

Australia’s Consumer Price Index (CPI) rose 0.6 per cent over the quarter in June, taking the annual rate of inflation up to 1.6 per cent from 1.3 per cent in March. That was slightly above market expectations for 0.5 per cent quarterly and 1.5 per cent annual results.

According to the ABS, about half of the 0.6 per cent quarterly increase in the headline rate was driven by a 3.5 per cent increase for the transport group (which comprises about 10 per cent of the total CPI basket), reflecting a sharp rise for automotive fuel (up 10.2 per cent) due to higher world oil prices.

Other categories that saw marked prices increases this quarter included medical and hospital services (up 2.6 per cent following the annual increase in private health insurance premiums in April) and international holiday, travel and accommodation (up 2.7 per cent). Items with significant price falls included fruit (down 4.1 per cent), electricity (down 1.7 per cent) and domestic holiday, travel and accommodation (down 1.5 per cent). The ABS noted that falls in administered prices (utilities, child care) continued to put downward pressure on the overall CPI.

Measures of underlying inflation remained subdued: the trimmed mean was up 1.6 per cent over the year, unchanged from the March quarter, while the weighted median was up 1.2 per cent and the CPI excluding volatiles rose by 1.5 per cent.

Why it matters:

Markets had been watching the June CPI numbers with next week’s RBA Board meeting front of mind: an unexpectedly weak reading would have spurred expectations of a possible policy change come 6 August. In the event, however, the second quarter inflation outcomes were slightly above consensus forecasts and as a result, expectations now are for the RBA to leave the cash rate unchanged this month.

Still, although the headline rate bounced back from March’s 1.3 per cent, it remained below both the RBA’s forecast of 1.75 per cent for the June quarter in May’s Statement on Monetary Policy and the bottom of the target band. Indeed, headline inflation has now been below the bottom of the band for 17 of the past 19 quarters and for nine quarters in a row, while the trimmed mean measure of inflation has been running at sub-two per cent for 14 consecutive quarters. On that basis, Australia remains mired in an extended period of lowflation.

What happened:

The number of dwellings approved in Australia fell by 1.2 per cent in June (seasonally adjusted). Approvals for private sector houses were up 0.4 per cent over the month, but that was more than offset by a 6.5 per cent slump in approvals for other private sector dwellings, reflecting the contraction in the apartment sector.

In annual terms, the ABS reported that total approvals were down more than 25 per cent from June 2018, with housing down almost 15 per cent and dwellings excluding housing plummeting by more than 39 per cent.

In trend terms, approvals were down 1.3 per cent over the month. That marks the 19th consecutive monthly decline in the trend rate, with the last positive print coming all the way back in November 2017.

In a related read on the housing market, data on new home sales (pdf) published by the Housing Industry Association showed sales up 0.8 per cent in the June quarter relative to the March quarter, marking the first quarterly increase since the final quarter of 2017. Over 2018-19 as whole, however, new home sales were down more than 12 per cent relative to the previous year.

Why it matters:

Although there are signs that the pace of decline in approvals has eased over recent months (particularly for houses, less so for apartments) the continued absolute decline in the numbers indicate that falling dwelling investment is set to be a headwind for the economy in the year ahead, along with a squeeze on employment in the construction sector.

That downbeat message is also consistent with a spate of negative businesses news from the construction sector this week, including Ralan Group’s entry into voluntary administration.

What happened:

According to CoreLogic, the combined capitals index of dwelling values rose by 0.1 per cent month on month in July while the national index was unchanged. Values were up in Sydney, Melbourne, Brisbane, Hobart and Darwin, with only Adelaide, Perth and Canberra recording falls. Values in Sydney and Melbourne are now 0.3 per cent and 0.4 per cent off their respective floors, while the rise in Brisbane was the first monthly increase recorded for that city since November 2018.

This was the first monthly increase in the combined capital cities index following 20 consecutive months of falling prices (and another two months of flat prices before that). It leaves dwelling values down 10.1 per cent from their peak, with values in Sydney 14.7 per cent below their previous high and Melbourne values down 10.8 per cent.

Why it matters:

After peaking at more than one per cent in December last year, the rate of monthly decline has been slowing ever since, with two RBA rate cuts, potential increases to households’ borrowing capacity due to APRA’s recent relaxation of lending requirements and the election outcome’s removal of the prospect of changes to the negative gearing regime all likely to have contributed to improving sentiment.

The bottoming of the market – and signs of an early recovery – should bring some useful relief to household balance sheets after a prolonged period of falling prices, and that in turn should be helpful for consumption prospects. But those same balance sheets remain burdened by high levels of debt which will remain a constraint on spending and leave consumers vulnerable to any adverse shocks. Meanwhile, the fall in approvals noted above is a reminder that the tough times for the construction sector certainly aren’t over.

What happened:

According to the RBA, total credit to the private sector was up 0.1 per cent in June over May (seasonally adjusted), and up 3.3 per cent over the year. Credit for housing grew 0.2 per cent month-on-month while personal and business credit were down 0.2 per cent and 0.1 per cent, respectively.

Why it matters:

Overall credit growth remains slow and there’s little sign of June’s rate cut stimulating any increase in activity. Monthly personal credit growth has been negative for some time as households continue to avoid personal loans but it’s notable that business credit has now been flat or falling for the past three months, with June’s monthly decline the first negative reading since January 2017.

What happened:

The Melbourne Institute released the 14th Annual Statistical Report of the Household, Income and Labour Dynamics in Australia (HILDA) Survey. HILDA is a nationally representative longitudinal survey of Australian households that has been tracking the lives of thousands of Australians each year since 20011. Because it is a longitudinal survey, it tracks the same households and individuals over time (adding to the sample children born to or adopted by sample members) and thereby offers a dynamic record of developments rather than a purely static snapshot.

The 2019 Statistical Report covers a wide range of issues including households and family life, the labour market and health expectations and health outcomes. For example, the data on commuting times received a lot of attention, with the survey showing that mean daily commuting times in Sydney have increased from 61 minutes in 2002 to 71 minutes in 2017.

The report also includes a detailed analysis of household economic wellbeing. Here, the data show that Australian real mean and median household disposable incomes grew strongly over the years between 2001 and 2009, and particularly strongly between 2003 and 2009, when both the mean and median incomes grew by around $3,000 per year2. Since 2009, however, growth has been much weaker: over the eight years to 2017, mean household income rose by only $3,156 while the median income in 2017 was actually $542 lower than it was in 2009.

HILDA also accounts for the potential impact of changes in household composition on these measures by tracking ‘equivalised’ income per person. Given the fact that there have been only small changes in household composition over this period, however, these numbers show the same pattern of strong growth in incomes between 2003 and 2009 and a subsequent levelling off.

By region, median incomes are highest in the Northern Territory (NT) and the Australian Capital Territory (ACT) and lowest in Tasmania and regional South Australia:

Over the period from 2012-13 to 2016-17, median incomes have declined in the ACT, Perth, regional WA, regional NSW and Adelaide but risen elsewhere:

As well as tracking shifts in the level of income, there is also an analysis of changes in its distribution. HILDA’s reported annual measures of equivalised disposable income show little change in income inequality between 2001 and 2017. For example, the ratio of income at the 90th percentile to the median income was virtually unchanged between 2001 (1.92) and 2017 (1.91), while the ratio of the median income to the 10th percentile was 2.12 in 2001 and 2.02 in 2017. Likewise, the Gini coefficient has remained between 0.29 and 0.31 over the entire 17 years of the HILDA survey, again, with very little difference between the 2001 and 2017 estimates.

Longer-term measures of income inequality paint a somewhat different picture. On the one hand, the Gini for five-year income is quite a bit lower than the annual Gini. On the other hand, this measure of the Gini has increased steadily from 2001-05 to 2013-17. 3

Why the discrepancy between the annual income and five-year income Gini coefficients? The HILDA report points to a modest decline in income mobility: the proportion of the bottom quintile of the income distribution still in that quintile one year later rose from 68.3 per cent in 2001-05 to 69.3 per cent in 2012-16, while the proportion of the top quintile still in that same quintile one year later rose from 70.7 per cent to 72.7 per cent over the same period. Greater income stability is good news at the top of the income distribution but unwelcome at the bottom, as it means that those in the latter group will then be more likely to have persistently low incomes.

The report also looks at intergenerational income mobility (that is, the extent to which individuals’ economic outcomes as adults depend on the economic outcomes of their parents) where the longitudinal nature of HILDA means that it is particularly well-suited for this kind of analysis. It takes individuals aged 15 – 17 in 2001 and compares their parental equivalised income in 2001 with their own household equivalised in income in 2017, now aged 32-34. This analysis finds a positive correlation between parental income and the income of children in later life: for example, of those whose parental income was in the bottom income quintile when they were a child in 2001, more than 34 per cent were themselves in the bottom quintile when they were an adult in 2017, compared to less than 10 per cent in the top quintile4. Two other interesting patterns found in the data are that this intergenerational correlation is stronger for female children than male children (although this finding was not robust for controlling for parental age), and the correlation between parent and child labour market earnings is higher for mothers than for fathers.

Finally, the 2019 HILDA report also presents data on trends in poverty, distinguishing between measures of relative and ‘anchored’ poverty. The relative income poverty rate is defined here as the share of the population with a household income below 50 per cent of median income while the anchored poverty rate is the share of the population with real income below the 2001 relative poverty line5. The share of the population below the relative poverty line has fluctuated over time, but the broad trend over the period of the HILDA survey has been downward, especially since 2007. The anchor rate of poverty has fallen much more dramatically. That said, the final year of the survey shows an increase in both relative and absolute measures.

Why it matters:

The HILDA data provides a fascinating look at the conditions of Australian households and the lives of Australian residents. The information on the economic wellbeing of households is of particular interest at present given the economic focus on the rate of improvement in living standards and importance of household consumption spending as a (currently lacklustre) driver of economic growth. Here, the survey shows that after growing strongly between 2003 and 2009, median real disposable household income has since stagnated. This provides supportive evidence both for the argument that economic performance in Australia has changed in the post-GFC period and for the closely related proposition that households are no longer enjoying the kind of income gains that they received during the opening decade of this century. If it is sustained, that’s a shift that is likely to have significant economic (and political) consequences.

. . . and what I’ve been following in the global economy

What happened:

The US Federal Open Market Committee (FOMC) announced on 31 July that it would lower the target range for the federal funds rate by 25bp to 2 - 2.25 per cent, with eight members of the FOMC voting in favour of the rate cut and two members voting against.

The accompanying statement said that the move was ‘In light of the implications of global developments for the economic outlook as well as muted inflation pressures’ and that in terms of future decisions, the Fed would ‘continue to monitor the implications of incoming information for the economic outlook and will act as appropriate to sustain the expansion’.

Fed Chair Jerome Powell provided a little more colour on the Fed’s thinking in comments (pdf) delivered in the post-announcement press conference when he stated that the rate cut was ‘intended to insure against downside risks from weak global growth and trade policy uncertainty; to help offset the effects these factors are currently having on the economy; and to promote a faster return of inflation to our symmetric two per cent objective.’ In the subsequent Q&A he also gave some guidance as to the likely future path of rates, saying of the rate cut ‘We’re thinking of it as essentially in the nature of a mid-cycle adjustment to policy’, later adding that he was contrasting the current move with the beginning of a lengthy cutting cycle, which was ‘not what we’re seeing now. That’s not our perspective now, or outlook.’

The FOMC also announced it would ‘conclude the reduction of its aggregate securities holdings in the System Open Market Account in August, two months earlier than previously indicated.’

Why it matters:

This was the first time that the Fed has cut rates since December 2008, when in response to the GFC it took the fed funds rate down to a target range of 0 – 0.25 per cent, where it then stayed for the next seven years. The Fed had embarked on a gradual process of monetary policy ‘normalisation’ with a 25bp increase in December 2015.

Eight more increases followed over the next four years, with the final rate hike coming in December 2018. That ‘normalisation’ process is now over but note that when the last Fed easing cycle started (in August 2007), the fed funds rate was at 5.25 per cent. The starting point for Powell’s ‘mid-cycle adjustment’ is, of course, much lower.

Normalisation has also involved shrinking the Fed’s balance sheet. As another element in its policy response to the financial crisis and its aftermath, the Fed bought more than US$3.5 trillion of bonds while implementing Quantitative Easing (QE). That saw the Fed’s balance sheet swell from about five per cent of GDP in 2007 to more than 25 per cent of GDP at its peak. The Fed started the process of shrinking its balance sheet in October 2017 in a policy that became known as ‘Quantitative Tightening’ (QT). By the end of July this year, it had contacted to a bit less than 18 per cent of GDP, and the FOMC announcement means that it will finish the normalisation process still well above its pre-crisis size.

With ‘normalisation’ done, it seems safe to say that the new normal for US monetary policy looks quite a bit different from the old normal.

It’s possible to interpret the Fed’s end-July rate cut in at least three ways.

• First, to simply take the FOMC at their word and view this as a combination of insurance against future weakening in the US economy and as a response to subdued inflation. It’s true that on some measures current US economic conditions look relatively robust and not in need of much additional policy support – second quarter US GDP growth beat consensus expectations with a decent 2.1 per cent (seasonally adjusted annualised rate) print, albeit down from Q1’s 3.1 per cent outcome, and the unemployment rate was still at a near-multidecade low of 3.7 per cent in June. But set against that there are signs that global weakness and trade tensions have been having an adverse impact on US investment and manufacturing, with recent manufacturing purchasing managers’ index (PMI) readings depicting a sector in near stagnation, for example. And US core inflation as captured by the core Personal Consumption Expenditure (PCE) price index was running at just 1.6 per cent in June with the headline PCE rate at 1.4 per cent, both below the Fed’s two per cent target6.
• Second, to see the rate cut as the Fed bowing to financial market pressure despite the FOMC’s own view that ‘sustained expansion of economic activity, strong labour market conditions, and inflation near the Committee's symmetric two percent objective are the most likely outcomes.’ Hence the two FOMC votes against the rate cut decision.
• Third, this could also be interpreted as the Fed succumbing to Presidential pressure for a more helpful policy stance.

If either the second or the third possibilities applied, however, the Fed’s actions appear to have done little to win over their target audiences. While markets had been expecting a 25bp cut in July, they had also been hoping for an indication that the Fed was set to deliver several more cuts over the course of the year. And although the Fed met the first part of these expectations, Powell’s comments ruling out ‘a lengthy cutting cycle’ served to disappoint both the US stock market and President Trump, who tweeted his displeasure (although he did concede he was happy to see the back of QT).

Finally, it’s worth noting that the Fed is not the only major central bank now struggling with how to manage financial market expectations. Last week, the ECB had its own turn at disappointing markets when it left policy rates unchanged, with the main refinancing rate left at zero and the deposit rate at minus 0.4 per cent. Outgoing ECB President Mario Draghi did say however that officials at the central bank are looking at new measures aimed at supporting the eurozone economy, which could include one or a combination of: a cut to the deposit rate, taking it further into negative territory; the introduction of a system of ‘tiered rates’ aimed at reducing the burden of negative rates on private lenders; and a resumption of the ECB’s own QE program, reversing the decision to cease bond purchases the bank made at the end of December last year. In the case of the eurozone, the economic data flow looks less ambiguous than the one facing the Fed: according to preliminary Eurostat data, the eurozone grew by just 0.2 per cent in the second quarter of this year, down from 0.4 per cent growth in Q1. That took annual growth down to 1.1 per cent from 1.2 per cent. And separate data showed the eurozone PMI for manufacturing hit its lowest level in more than six years in July.

What happened:

President Trump said that the United States would start imposing an additional tariff of 10 per cent on a further US$300 billion of Chinese imports from the beginning of September. The move came just after trade talks between Washington and Beijing had resumed this week, with a short meeting in Shanghai. The meeting was the first after both sides reached a ceasefire in their trade war during last month’s G20 meeting in Osaka. The next round of talks is scheduled for September.

Why it matters:

The new measures would see Washington impose tariffs on essentially all Chinese imports into the United States, including a range of consumer products. It marks another significant escalation in the bilateral trade conflict between the two economies and is likely to increase the uncertainty currently damaging world trade, manufacturing and investment sentiment.

US-China trade tensions are one of the key risks hanging over the global economic outlook, meaning that financial markets and world economy watchers closely follow every meeting – and every Presidential tweet. Despite the various ups, downs and breakdowns in the negotiations however, the broad picture remains unchanged. As we’ve noted before, if the acceptable end game for both sides is a modest agreement that delivers limited but meaningful Chinese concessions in the form of more purchases of US goods and services and some domestic liberalisation, then that looks achievable, if still vulnerable to the shifting political priorities in both capitals. But if the end game is supposed to deliver a much more fundamental restructuring of the whole Chinese economic model, then any durable resolution likely remains as far away as ever. Meanwhile, world trade growth remains decidedly weak, global supply chains remain under pressure to reconfigure themselves in order to adapt to the changing geopolitical order, and global manufacturing and investment spending has been suffering, including in the United States and China (although see also the Gavyn Davies piece linked to in the reading list below).

What happened:

The British pound has been sliding in foreign exchange markets this week.

Why it matters:

Last week’s note looked at the arrival of a new UK Prime Minister and the impact of this on the chances of a no-deal Brexit, arguing that the probability of the latter had now increased. The fall in the pound over the last few days indicates that this take on the shift in risk probabilities is shared by the currency markets, at least for now.

What I’ve been reading: articles and essays

The August issue of AFR magazine has an interesting profile of RBA governor Philip Lowe.

The RBA has posted the papers from its April 2019 conference on low wage growth. Lots of content here, but the opening paper (pdf) from Natasha Cassidy provides a nice snapshot of the recent trends in Australian wage growth that the conference was seeking to explain (a relatively abrupt, broad-based and persistent slowdown in the rate of growth of nominal wages, particularly between 2013 and 2016) while the concluding discussion (pdf) offers a quick summary and overview of the various arguments (possible ‘suspects’ include the economic cycle, a changing industrial landscape and a shift in the balance of power between employees and employers, but with insufficient evidence to single out any one cause ‘beyond reasonable doubt’.)

Alan Oxley and Innes Willox propose decoupling trade and security in Australia’s foreign policy.

An ANZ bluenote on Australia’s progress with water and energy efficiency use draws on the latest environmental-economic accounts from the ABS (showing that the energy and greenhouse gas emission intensity of output has declined but water intensity has increased) to point out that efficiency gains to date have not been enough to offset the greater resource use required by population growth.

Two pieces related to the short comment above on the fall in sterling: Schroders’ senior European economist explains why the benefits of a weaker currency to the UK economy have so far turned out to be relatively limited and John Authers on BoJo versus the market.

Stiglitz, Durand and Fitoussi make the case for the OECD-hosted High Level Group on the Measurement to Economic Performance and Social Progress and its quest to develop better datasets and tools to narrow the gap between expert views and popular perceptions.

Chris Dillow argues that economic liberalisation may have been bad for productivity.

Bloomberg graphics presents a series of paired charts arguing that globalization isn’t dying, it’s just evolving.

Also from Bloomberg, the Hajiyeva case and London’s push to curb the flow of dirty money, including the ‘Azerbaijani Laundromat’.

Gavyn Davies reckons that the US-China trade wars have wrecked more damage on the European and US economies than they have on the Chinese one.

The IMF on a stalling recovery in Latin America plus six charts on boosting growth in Brazil.

Institutional Investor looks at the world of risky loans.

1 The first survey in 2001 involved the participation of 7,682 households with 19,914 residents (although only those aged 15 or over are interviewed). The survey has since grown, including for example through a sample top-up (of more than 2,000 households and more than 4,000 respondents) to capture immigrants who arrived in Australia between 2001 and 2011.

2 The mean is the average income across households. The median is the middle data point in household incomes sorted from lowest to highest, such that half of all data points are below the median and half are above it. When incomes are distributed unevenly, the conventional assumption is that the median will be a better measure of the ‘typical’ income. That’s because the mean will be influenced by outliers, such as a relatively small number of very high incomes. The classic illustration of this effect is to imagine a room containing nine people, each with a ‘normal’ net worth, and tenth person who is a multibillionaire (Bill Gates has been the popular choice in recent years). The average or mean net worth in the room is of course incredibly high thanks to the presence of Bill, even though the ‘typical’ net worth in the room will be a tiny fraction of that. Taking the median gets around this shortcoming.

3 Five-year income is the sum of real annual equivalised income over the corresponding five years. This ‘smooths out’ annual fluctuations in income. Since longer-term income tends to be more evenly distributed (have a lower Gini) than annual income, the latter can be a misleading indicator of the ‘real’ level of income inequality, since in general we will be more concerned about lifetime income than income in just one year.

4 Data for one-year income. For three-year income the corresponding ratios were 37.6 per cent and 5.5 per cent.

5 Relative poverty on this measure tracks the ability to sustain a ‘typical’ lifestyle while the anchored poverty line is focused on the real value of the purchasing power of the poverty line.

6 Formally, the Fed targets the overall PCE inflation index, but it also tracks the core measure, which excludes food and energy prices, as a superior measure of underlying inflation.