Jobs data not as strong as it seems...

In the AFR today I write about what an incredible week it was for markets. The fourth-largest monthly decline in unemployment on record and yet the Aussie dollar fell yesterday from US76.1 to US75.5 cents (or from JPY84.3 to JPY83.3) care of a prudently dovish Martin Place.

Hours earlier, a more hawkish US Federal Reserve triggered a 10-basis-point jump in long-term interest rates only for this to be reversed out one day later as markets digested its communications and higher than expected dole payments. As National Australia Bank cautioned on Friday, “the road to recovery isn’t straightforward, with US unemployment claims rising last week and Britain’s infection rate increasing sharply, despite the vaccines”.

While the Treasury and the Reserve Bank of Australia can claim much credit for our exceptional post-crisis recovery, the drop in the unemployment rate from 5.48% in April to 5.07% in May, just above its January 2020 level, conceals much more complexity than the market understands.

As Commonwealth Bank’s Gareth Aird has highlighted, the official statistician publishes a broader measure of employment, called the Labour Market Account, which includes non-resident workers. And this tells a very different story.

Instead of the total number of employed persons in Australia being 1 per cent above its pre-COVID-19 2020 peak, the broader Labour Market Account finds that in March 2021 employment was actually 2.1% below its level one year prior. This is being driven by the exodus of non-resident workers out of Australia. Between December 2019 and March 2021, the number of non-resident workers slumped by 333,900 persons.

And yet these non-resident jobs, which were not a part of the more commonly cited Labour Force Survey, have to be filled. So employers have been hiring locals. We have effectively had 333,900 new jobs that were not in the Labour Force Survey start shifting into it (those jobs that are not placed are picked up by the vacancies survey).

This helps explain the sharp reduction in the unemployment rate, as residents take this newly available work, and the soaring vacancy rate as employers advertise unfilled spots. It also explains why the jobless rate in regional markets, where many non-residents worked, has declined much more quickly to be below pre-pandemic levels in contrast to the cities where the unemployment rate remains above this threshold.

The problem for the RBA is that as most of the population is vaccinated over time, borders will gradually reopen. Governments are already creating new pathways for foreign students to return and migrants will follow quickly thereafter. Over the next few years, much of the lost non-resident labour supply could return. Australia is, after all, now one of the most attractive destinations for tourists, students, and skilled migrants, given our economic and health performance since the crisis. A cheap Aussie dollar only amplifies our relative appeal.

If one assumes the 333,900 non-resident jobs are all placed with locals who were looking for work (and not folks who are sucked into the labour market), this has the effect of temporarily reducing our unemployment rate by 2.4% points. Put another way, if borders open and non-resident workers return and take these jobs, the unemployment rate would increase from 5.1% to 7.5%. Although this is a very crude upper bound, it gives a sense of the influence of the non-resident worker exodus on the jobless rate.

On Thursday the RBA’s governor, Phil Lowe, touched on these complexities in a thoughtful speech on how the central bank plans to get wages and inflation back on track.

The RBA knows it is in unchartered forecasting territory, which will include navigating the reopening of borders and the sharp expansion in population growth and labour supply. This could mean that any future bump in wages growth is transitory as the negative labour supply shock wrought by COVID-19 reverses out.

In Lowe’s speech, which was accompanied by a raft of important new RBA research papers, he delivered several messages. Perhaps the most significant was that despite the recovery being “much stronger” than the RBA had anticipated, “wages growth and inflation remain subdued”. In fact, we are getting the wages and inflation results that the RBA was predicting would arise assuming far weaker growth and significantly higher unemployment.

“The Wage Price Index increased by just 1.5% over the past year, with wages growth slow in the private and public sectors,” Lowe said. This is less than half the 3 to 4% pace the RBA needs to normalise inflation back into its target 2 to 3% band. “And it is noteworthy that even in those pockets where firms are finding it hardest to hire workers, wage increases are mostly modest,” Lowe continued.

Lowe argued that Australia’s weak wage growth partly reflects the reluctance of businesses to raise fixed-costs after numerous macro headwinds, including the 2011 to 2013 period when the Aussie dollar soared above 100 US cents, rendering many exporters and import-competing firms uncompetitive, and the more recent pandemic, which one way or another threatened most companies existentially.

The RBA’s analysis reveals many employers are “relying on non-wage strategies to retain and attract staff”. “Some are also adopting a ‘wait and ration’ approach: wait until labour market conditions ease, perhaps when the borders reopen, and until then, ration output,” Lowe said.

“For some, this is a better option than paying higher wages and driving up their own cost base. This is especially so if increases in the cost base are difficult to reverse later on, there is a reluctance to increase prices, and the business expects labour market conditions to ease before too long. By waiting and rationing, firms can avoid entrenching a higher cost structure in response to a problem that might be only temporary.”

Our macro strategist Kieran Davies says this reminds him of the dynamic during the global financial crisis when Australian firms did not shed large numbers of staff because the cost of rehiring them would have been too great.

It is a rational approach if you believe borders will open up over the next year or two. It also sheds light on the disconnect between the wages and inflation data and Australia’s increasingly tight labour market, which will loosen up as borders become more porous.

Finally, it reinforces the point that the RBA’s forecasting models, which have consistently overestimated wages and inflation for years, are likely to perform poorly in this unprecedented environment. This is why Lowe and his deputy governor Guy Debelle have emphasised that Martin Place will predicate policy decisions on what they consider to be credible wages and inflation data, rather than volatile jobs numbers and/or rubbery forecasts of the future. And this data will have to convince them that wages are rising sustainably at 3 to 4 per cent annually.

Despite this commitment to “nowcasting”, the RBA’s stimulus is still being slowly unwound. On 30 June they will stop lending banks ultra-cheap, 3-year money at a 0.1 per cent rate, which will push mortgage rates up as banks eventually replace this $200 billion of funding with more expensive wholesale debt. And in July the RBA will not extend its 3-year yield curve target of 0.1 per cent from the April 2024 bond to the November 2024 security, signalling that the first-rate rises will materialise around this time.

The one remaining tool the RBA has have left to furnish further stimulus is their government bond purchase (or quantitative easing) program, which places downward pressure on long-term interest rates that has slowed the ascent of our exchange rate notwithstanding extremely elevated commodity prices.

According to Bill Evans, the Aussie dollar should be trading between US85 and US90 cents right now based on Westpac’s exchange rate model. “The only thing we can finger that explains the gap between Aussie dollar’s fair value and its current level around US75 cents is the RBA’s QE program,” Evans says.

On Thursday Lowe said that “the RBA's bond purchase program is one of the factors underpinning the accommodative conditions necessary for our economic recovery”, reiterating that it would be “premature” to consider curtailing those purchases when peer central banks, like the US Federal Reserve and the European Central Bank, are expected to continue their own QE programs through 2022. On Tuesday the RBA's board noted that its QE commitments to date lag the rest of the world on a relative basis, implying comparatively inferior stimulus.

“The key consideration in our decision is how the RBA can best support the ongoing recovery of the economy,” Lowe said. “The Board wants to see the recent recovery transition into strong and durable economic growth, with low unemployment and faster growth in wages than we have seen recently. Over time, this will help achieve the inflation target.”

Lowe lent further weight to what has become the prevailing market consensus, prompted by The Australian Financial Review’s John Kehoe, for the RBA’s QE program to evolve in September to an open-ended, $5 billion per week initiative that is reviewed quarterly and calibrated based on hard evidence about wages and inflation. This view is now backed by most banks including Westpac, ANZ, Goldman Sachs, HSBC, RBC, Nomura, UBS, and Deutsche Bank.

The RBA's resolute commitment to maintaining its stimulus to help Australians secure full employment characterised by sustainable wages growth and inflation within the target band is a key reason why the Aussie dollar declined against other currencies despite Thursday's stunning jobs data.

And if there was any doubt as to what the RBA thought about the efficacy of these policies, it released a rich array of new research papers that evaluated their performance. On the question of QE, the RBA’s economists “estimate that the program has reduced longer-term Australian Government Security (AGS) yields by around 30 basis points and lowered the spread of state and territory bond yields to AGS yields by 5 to 10 basis points, relative to where they would otherwise have been”. They further found that “the bond purchase program has not had any substantial negative impact on the functioning of government bond markets”. Indeed, most market experts think that the RBA’s purchases of 5-year to 10-year government bonds have substantially enhanced liquidity with no evidence that it is crowding out other participants. 

Access Coolabah's intellectual edge

With the biggest team in investment-grade Australian fixed-income, Coolabah Capital Investments publishes unique insights and research on markets and macroeconomics from around the world overlaid leveraging its 14 analysts and 5 portfolio managers. Click the ‘FOLLOW’ button below for more of our insights.