News & analysis

Today’s jobless claims data from the US saw another dramatic rise in the number of unemployed persons claiming insurance benefits.

The 5.245m number of claims registered in the week ending April 11th has seen the aggregate number of job losses, if this is a true measure, erase the employment gains made since the global financial crisis. The data followed a 6.62m increase in claims the week prior, pushing the four-week sum to 22m filings compared to the 21.5m jobs added during the economic recovery beginning June 2009.

The US dollar has gone mildly bid in markets following the release as risk sentiment continues to drive market pricing. However, despite today’s data highlighting a dramatic collapse in the US labour market, with the advanced seasonally adjusted unemployment rate rising to a record high of 8.2% for the week ending April 4th, it isn’t necessarily an accurate representation of the real economy. Below, we highlight some of the reasons for this which include; the loss of output from those leaving the job market altogether therefore not claiming unemployment benefits, system failures in states, backlogs in processing and the impact of the CARES act.

State data for week ending April 4th:

  • Georgia (+256,312), Michigan (+84,219), Arizona (+43,488), Texas (+38,982), and Virginia (+34,872) saw the largest rise in claims.
  • California (-139,511), Pennsylvania (-127,037), Florida (-58,599), Ohio (-48,097), and Massachusetts (-41,776) saw the largest falls in claims.

A note on US Jobless claims in general

The US Jobless claims figure has been one of the most contentious data points during the coronavirus outbreak. While the data is one of the timeliest metrics to assess the impact lockdown measures are having on the US labour market, in turn allowing markets to gauge the effects on consumption and therefore growth, it is subject to a number of quirks. The quirks make it hard for economists to filter through the noise and truly measure the economic impact of social distancing measures in the US via the labour market, but given the lack of timely data markets will still use this as their best nowcast of the US economy. This means that initial jobless claims are likely to continue having an elevated market impact as more accurate unemployment data is compiled.

Last week, the initial jobless claims data release saw the number of people claiming unemployment benefits in the week ending April 4th dip trivially to 6,606K from an upwardly revised 6,867K the week prior. The dip in jobless claims was concentrated in Pennsylvania (-143K), Florida (-108K) and Massachusetts (-103K), but recent reports in the Washington Post highlight that the benefits system in states like Florida has led to the state underreporting the true number of claims. Rebecca Vallas reported that the red tape surrounding the state claims system was deliberately designed by the former Governor Rick Scott to “make it harder for people to get and keep benefits so the unemployment numbers were low”. This is supposedly just the tip of the iceberg.

System failures nationally have arguably suppressed the true number of new unemployment claims. Reports of system crashes in New York, Michigan and Nevada are also backed up by reports of increased server capacity nationally and extended call center hours beginning next week. The influx of claims and the current backlog suggests that the initial claims data will remain elevated for some time to come as opposed to seeing the data subdue and a corresponding rise in continuing claims. Additionally, the CARES act will now see more claims fall under the official data stemming from unemployment benefits being paid to self-employed and gig economy workers. With the widening criteria of claims, expect the nominal number to remain elevated in the numerous millions for some weeks to follow.

While the multitude of factors in play makes it harder to assess the true economic damage, further drawbacks are found in the fact that the claims data doesn’t measure the loss of economic activity from those leaving the labour market all together.

In the March household survey, approximately 60% of workers that lost their job were counted as no longer in the labour market because they didn’t want a job or were not looking. Additionally, the structural contraction in the economy is difficult to assess due to the amount of temporary claims crowding out the true structural unemployment rate. Markets know the short-term contraction in the economy is set to be sharp but much of this can be recovered once containment measures are relaxed. The concern is what the longer-term impact to the economy is and to measure that, and in turn the effectiveness of the stimulus measures to counteract it, markets will need to see the noise around the labour market data to subside. While containment policies are in place, the data will continue to be marred with noise and only reflect the short-term economic impact, but it remains the best and most timely measure of economic activity for now.

Eurozone Industrial Production data fails to move euro

This morning’s eurozone industrial output data from February included a sharp contraction of 1.9% on a year-on-year basis and a 0.1% contraction month-on-month, with a decrease in capital and durable goods causing much of the slump. The month-on-month slump is negligible due to the increase in intermediate and non-durable consumer goods, as well as energy production.

The data was already hampered in February as the virus had China under lockdown, causing many disruptions in supply chains of trading partners, including the eurozone. With eurozone countries following nationwide lockdowns in March, a sharp downturn in the next data release is a given. Reasonably encouraging purchasing managers’ indices from March suggested on the surface the slowdown was particularly sharp, but the PMIs were subject to noise from the inverse effect of the supply delivery times.

With some eurozone countries looking at options to gradually revive the economy again, an important question that arises is when factories and companies will resume production. Germany has already announced plans to ease some of containment measures at the beginning of May and will reopen some smaller shops, while Spain will allow workers in manufacturing, construction and some services to return to work from this week onwards. The rest of the Spanish population remains in lockdown. The moment at which countries resume production will be the largest factor in assessing what the lowest point of production will be and will pave the way for a recovery in the eurozone industry.

The reaction in FX markets to the data remains limited…

As markets are mainly concerned with the shift in risk sentiment that has appeared since this week’s collapse in oil markets and renewed safe haven demand for the US dollar. The euro has weakened against the dollar throughout the day, including the moments after the data release, but remained relatively stable against Sterling, signalling that this story is a matter of dollar strength rather than euro weakness.

Looking ahead, industrial production data for March and April, combined with the timing of the easing of containment measures will be key in determining the bottom eurozone’s industrial downturn.

 

Eurozone industrial production vs surveyed values

 

Eurozone industrial production February 2020 (Covid Crisis) vs January 2009 (GFC)

 

Mexico’s rating erosion will make it harder for the peso to recover

The ratings agency Fitch downgraded Mexican sovereign debt to BBB- from BBB yesterday, just one notch above speculative investment grade, with a stable outlook. The downgrade comes after the coronavirus crisis has dampened the already grim outlook of the Mexican economy and its public debt outlook. The agency now foresees an economic collapse of at least 4% in 2020, while pushing prospects of recovery back to 2021. Despite the fact that the government has not enacted any additional fiscal measures to rescue the economy amid the coronavirus crisis, the economic fallout could represent an increase of the primary deficit as percentage of GDP of some 2.5pp to 4.4%. In turn, this could push the debt-to-GDP ratio above the 50%, a record not seen since the 1980s. Fitch is the second major agency to cut Mexico’s sovereign debt rating in less than a month, after S&P downgraded it to two notches above junk at the end of March.

The country is extremely sensitive to further downward revisions by major agencies despite AMLO’s austerity stance, as structural growth could be impacted for years to come. According to the OECD’s PPP standard, the Mexican peso is currently undervalued by nearly 160%, while the currency has lost over ¼ of its nominal value in less than two months.

Even if USD strength starts to ease in the following quarters worldwide, poor domestic fundamentals will hardly help the peso to recover towards its pre-coronavirus levels in the short-to-medium term horizon.

 

Mexican debt-to-GDP ratio is set to jump to above multi-decade levels amid coronavirus economic collapse

 

 

Authors: 
Simon Harvey, FX Market Analyst
Olivia Alvarez Mendez, FX Market Analyst
Ima Sammani, FX Market Analyst

 

 

DISCLAIMER: This information has been prepared by Monex Europe Limited, an execution-only service provider. The material is for general information purposes only, and does not take into account your personal circumstances or objectives. Nothing in this material is, or should be considered to be, financial, investment or other advice on which reliance should be placed. No representation or warranty is given as to the accuracy or completeness of this information. No opinion given in the material constitutes a recommendation by Monex Europe Limited or the author that any particular transaction or investment strategy is suitable for any specific person. The material has not been prepared in accordance with legal requirements designed to promote the independence of investment research, it is not subject to any prohibition on dealing ahead of the dissemination of investment research and as such is considered to be a marketing communication.