Thursday, September 3, 2015

3/9/15: BRIC Servies PMI: August


BRIC Services PMIs are out, so here is the summary of latest changes:


  • Brazil Services PMI rose to 44.8 in August from July's 39.1. This marks seventh consecutive month of Services reading below 50.0. July fire was horrific and cam on top of abysmal 39.9 reading in June. August reading is still below 50, but is now the second highest in 7 months. In 12 months from August 2014, the index posted above 50 reading in only two months. This confirms deep-running recessionary dynamic in Brazil's economy.
  • Russia Services PMI came in at 49.1, below 50.0 and down on 51.6 in July. Details covered here. As noted, this is a disappointing reading signalling weak contraction in the Services side of the Russian economy, compounding sharper contraction in Manufacturing.
  • China Services PMI posted 51.5 in August, down from 53.8 in July. This index has never reached below 50.0 in its history, so 51.5 is a weak reading overall. In fact, this is the slowest pace of expansion signalled by the Services sector since July 2014. Taken together with rather sharp ongoing contraction in Manufacturing, the Services PMI for China suggests that current rate of economic growth in the country is well below the target of 7% annual rate of growth targeted by the Chinese Government. Most likely, stripping out financials (supported by the Government aggressive intervention in the stock markets in recent months), China's Services PMI would be in much worse shape.
  • India Services PMI came in at 51.8 - signalling acceleration in growth from 50.8 in July. This marks second consecutive month of above 50.0 readings for this index. 

Some Markit comments on the above:

For China: "The Caixin China General Services Business Activity Index …signalled the slowest increase in activity in the current 13-month sequence of expansion. Total new order growth at service sector companies also weakened in August, with the latest increase in new business the slowest seen in just over a year. According to anecdotal evidence, relatively subdued market conditions had dampened client demand in the latest survey period."

For India: "The seasonally adjusted Nikkei Services Business Activity Index [pointed] to a faster, although modest, expansion in output. Activity growth was recorded in three of the six monitored categories, namely, Hotels & Restaurants, Post & Telecommunication and ‘Other Services’. Leading services activity to increase was a further rise in incoming new work. According to panellists, higher demand coupled with capacity improvements and increased marketing all had contributed to the latest expansion in new business. Despite quickening since July, the rate of growth was moderate. While manufacturing new work rose at a weaker pace than in July, growth nevertheless outpaced that seen at their services counterparts."



CONCLUSION: BRIC economies are suffering from the twin effects on internal structural weaknesses (domestic markets and policies) and from lack of external demand growth. These leading emerging market economies are now solidly in a recessionary dynamic (Brazil and Russia), and near hard landing territory (China). India continues to break trend with the rest of BRICs, although this remains to be seen how long its economy can sustain this decoupling. 

3/9/15: Russian Manufacturing, Services & Composite PMIs: August


Russia PMI data for Services, Manufacturing and Composite posted sub-50 performance across all three indicators in August, returning the economy back to where it was around June 2015, and erasing the fragile expectations of stabilisation that were based on July data.

As noted in my analysis of BRICs manufacturing PMIs earlier (link here):

Russia Manufacturing PMI fell to 47.9 from 48.3 in July, marking 9th consecutive month of sub-50 readings and worst performance in the sector since May 2015. August move effectively demolished previous expectations of stabilisation in Manufacturing sector in Russia.

Per Markit release: "Operating conditions in the Russian manufacturing sector continued to deteriorate during August amid reports of a deterioration in the economic environment. Output was little changed, while new orders and employment both fell to the greatest degrees since May. Notably, a depreciation in the Russian rouble against the US dollar led to a sharp and accelerated increase in average input prices by raising the cost of imported goods. …The net effect was a decline in demand and a drop off in levels of incoming new business."

Meanwhile, Services PMI posted a disappointing decline from 51.6 in July to 49.1 in August, pushing the index below 50 mark once again. The index fell to its lowest level for the period covering last 5 months.

Per Markit: "The Russian service sector registered a slight fall in business activity during August as incoming new orders were barely changed and excess resources remained evident. Backlogs of work were again cut sharply, placing further downward pressure on staffing levels… Undermining service sector activity was a general lack of growth in incoming new business. Latest data showed that new work was only marginally higher, with companies bemoaning a lack of funds at clients amid evidence of a challenging economic environment.


With booth Manufacturing and Services down, Composite PMI for Russia fell below 50.0 marker in August, reaching 49.3 against 50.9 in August. This marks the second month in the last 3 months of sub-50 readings and August Composite PMI level is at the lowest levels since April 2015.

SUMMARY: As I noted consistently in the past, any sign of stabilisation in Russian economy coming on foot of disappointing 1H 2015 will require several confirmations before we can call a switch in the growth trend. This confirmation (on foot of July upside performance) did not arrive to-date.

3/9/15: BRIC Manufacturing PMIs: August

BRICs manufacturing PMIs signalled continued worsening in growth conditions in world's largest emerging markets.


  • Brazil Manufacturing PMI fell to an abysmally low 45.8 in August compared to already poor 47.2 in July. This marks the fastest rate of decline in manufacturing activity in the economy since September 2011 and the 7th consecutive month of sub-50 readings in the index.
  • Russia Manufacturing PMI fell to 47.9 from 48.3 in July, marking 9th consecutive month of sub-50 readings and worst performance in the sector since May 2015. August move effectively demolished previous expectations of stabilisation in Manufacturing sector in Russia. In my previous posts on the subject I have consistently noted that early signs of such stabilisation were yet to be fully confirmed and we will have to wait until we see Services PMI for Russia for more analysis.
  • India Manufacturing PMI continued above-50 trend performance in August, although the index did fall to 52.3 from 52.7 in July. Statistically-speaking, 51.5 for the Indian economy is consistent with moderate growth. Overall, Indian Manufacturing PMIs have now been in continuous expansion territory over 22 consecutive months.
  • China Manufacturing PMI came in at a disappointing 47.3 in August, down on already poor 47.8 in July, marking 6th consecutive month of contraction in the sector. Overall, August reading is the lowest since March 2009. The trend suggests the economy is nowhere neat the target of 7% annual growth rate targeted by the Beijing officials.



Summary view: Overall, BRIC Manufacturing PMIs signalled deepening of the ongoing economic growth slowdown in the largest emerging economies. We will need to wait for the analysis of Services and Composite PMIs to confirm this, but August has been a disappointing month for the prospects of global growth recovery.

Wednesday, September 2, 2015

2/9/15: House prices, rents and Irish demographics


Based on latest data from CSO, Dublin has gained disproportionately in population compared to the rest of the country in 12 months through April 2015. At the top level, Dublin population rose 30,700 in 2015 compared to overall state population rise of 25,800, which implies that ex-Dublin, the country lost population at the rate of 4,900. Dublin v rest of the state population changes were even more dramatic when one considers age distribution.

  • Population aged 0-19 years rose in Dublin by 15,700 in 2015. The same group numbers increased state-wide by 19,600. Which means ex-Dublin, younger population rose only 3,900 against Dublin's 15,700. This highlights family formation dynamics in Dublin as opposed to the rest of the country.
  • Younger working-age population (age 20-29) fell in Dublin by 5,700 in 2015. Across the country, the decline was 25,400. Which means that ex-Dublin, younger working-age population has declined by 19,700.
  • The main cohort of working-age population (30-64 years old) has risen in Dublin by 15,700 in 2015 compared to 2014. Across the country the increase was 14,100, which implies that ex-Dublin, the state lost 1,600 adults in the main working age cohort.
  • Older population dynamics were also in favour of Dublin. In 12 months through April 2015, population aged 65 and older rose in Dublin by 4,900 and it was up nationwide by 25,800. Which means that ex-Dublin, older age population rose 20,900.




The above dynamics suggest demographic support for rising property prices and rents in Dublin, as was suggested by some analysts. This may or may not be so, since population dynamics work over years, not in simple y/y terms. So let's take a look at relative changes in Dublin population compared to the rest of the state:


Dublin population, as % of state-wide population of core home ownership demographic that transacts on the purchasing side of the market (30-64 year olds) has been rising overall since falling to 27.5% in 2010 and is currently standing at 28.9%, still below 29.6% high in 1996. But the rental demographic of 20-29 year olds has shown different dynamics, reaching period trough in 2013 (at 32.3%) and rising since then to 33.1% in 2015, a level consistent with 2011. Neither suggests huge uplift in demand for rentals or owner-occupied homes.

Of course, these are cohorts of 2015. Back in 2007, when house loans were last available in plentiful supply, large share of purchasers demographic today was… err… renters. And absent credit for house purchases, as this demographic moved into purchasing age, they stayed renters.  This would suggest increased pressures on rents. But the picture is more messed up by the losses of population share in 20-29 years old group, most likely due to emigration. Longer tenure for children staying in parents' homes also should have held rents back. In other words, our traditional view of demographic distribution of buyers v renters has been messed up by the sheer duration of the current crisis.

Take 2010-2015 period. Over that time, Dublin population of 20-29 year olds fell 61,000 and across the rest of the country, this cohort numbers fell 123,000. But cohort of 30-64 year olds rose in Dublin by 60,900 against a rise of 47,600 in the rest of the country. Demand based on demographics, therefore, suggests swing from renting toward purchasing. Similar picture repeats if we take 2011-2015 cumulative changes and 2012-2015 and 2013-2015. And, still, rents were rising.

What is more mysterious is that overall working age population has been relatively mildly altered in recent years. In Dublin, working age cohort (20-64 year olds) has grown by just 10,000 in 2014-2015 period and it is up only 8,000 on 2012. The cohort is still down 13,300 on 2008. Across the country, ex-Dublin, things were much worse: since 2008 the cohort fell 62,200 and compared to 2014 it is now down 21,300. In simple terms, unless children and retirees are buying homes, there shouldn't be any dramatic uplifts in demand for property in Dublin, and most certainly outside Dublin.

Which goes to say that any claims about actual demand (based on numbers of potential renters and buyers) are a bit strained. In 2008, there were 821,200 people of age 20-64 living in Dublin. Today there are 807,900. Unemployment rose over that time too. So where is that tremendous growth in demand coming from to push property prices up? Property prices in Dublin hit a period trough in 2012. Since then, there has been a net increase of just 8,000 in 20-64 year olds cohort living in Dublin. Again, where is that spiking in demand coming from?

In my view, simple demographics do not explain Irish property prices uptick from crisis lows. Speculation and latent surplus of savings in a relatively small category of Irish residents (and ex-past), plus re-distribution within cohorts between renters and buyers are the main drivers on demand side of the equation. Supply side also contributes significantly to prices uptick. And beyond that, there has to be significant behavioural component: our addiction to property - despite all the hopes of Dublin 'planners' for an Amsterdam on the Liffey - has not gone away over the years of the crisis. The first, initial shock to the economy has had an effect of scaring us put of the markets for property. Negative equity contributed more downward momentum. But once we learned to live with our fears, we simply decided o turn back to our old model of family investment: bricks and mortar. 

Sunday, August 30, 2015

30/8/15: Migration & Changes in Irish Population: Working Age Population


In two previous posts, I looked at Irish migration and population changes data from the point of:
- Top level analysis of migration and natural changes in population; and
- Migration trends by nationality.

Continuing with analysis of population data from CSO released earlier this week, let's take a look at the age composition of population.

In what follows, I define two key categories within our population:

  • Working age group - population aged 20 years through 64 years. This is an approximate definition, and I prefer it to including 15-20 year olds into it, primarily because it allows for more accurate reflection of numbers in full time education. There are many caveats applicable here, so take the approximation for what it is - indicative, rather than definitive.
  • Non-working age population (rest of population). Again, that is not to say that younger students do not work (at least part-time) or that people beyond 65 years of age do not work. Some do. Majority do not. When many do work, they work less hours than is required to sustain independent living, so they still rely on either pensions or social transfers or family transfers or any permutations of the three to sustain themselves.

In simple terms, mindful of all caveats, etc, a ratio of working age population to non-working age population tells so a bit about how high is the dependency weight in the society due to age distribution in population. Lower ratio means fewer working age adults having to sustain themselves and non-working age people. By sustain I mean economically sustain - by working and adding value in the economy.


Chart below shows distribution of changes in working age population and non-working age population in y/y and cumulatively from 2008:


What stands out in the chart?

  1. Working age population overall has been in decline since 2010. In cumulative terms, number of working age adults has fallen 2.7% on 2008 level by April 2015 - a decline of 75,500. In 2015, the rate of decline was 0.4% - more moderate than in three previous years, but still steeper than 2010-2011 average.
  2. Non-working age population remains on the rise for every year covered by the CSO series and the rate of increase in 2015 (at 2%) was the highest since 2010. Overall, over 2008-2015 period, non-working age population those 13.3% or 225,900.
  3. The gap between working age population and non-working age population is now at 798,400 - the worst reading in series history and 301,400 worse than in 2008.


As the result of the above trends, ratio of working age population to non-working age population continued to fall precipitously in 2015:


In 2015, the ratio of working age population to non-working age population was 1.42 - meaning that for each non-working age person, there were 1.42 working age adults. This does not correct for the working age adults who are not in the labour force as well as for the unemployed. The best performance year in this metric was 2007 when the ratio was 1.66. In other words, in 2015, there were 24 fewer working age adults per each 100 non-working age persons than in 2007.

Saturday, August 29, 2015

29/8/15: Migration & Natural Changes in Irish Population: Migration by Nationality


Having looked in the previous post at top level data for population changes in Ireland reported by CSO, now let's take a look at composition of migrants flows by nationality. This is going to be charts-heavy.

Let's start with immigration flows. Chart below shows Immigration into Ireland over the recent years:


Several interesting aspects of this jump out:

  1. There has been a significant increase of inflows of people from the 'Rest of the World' (ex-EU). Numbers of those coming into Ireland from outside the EU are up at 30,400 in 2015 from 25,500 in 2014. Pre-2015, average annual inflows of immigrants from outside the EU was 15,722, so last year things were pretty much ahead of the average for the third year in a row. Much of this is probably driven by big hiring numbers from multinationals which are increasingly moving their EMEA and MENA operations into Ireland. 
  2. There has been a small uptick in the number of new comers from the Accession States (EU12), the numbers of which rose to 12,800 in 2015 compared to 10,000 in 2014. This is the second highest inflow rate since the start of the crisis. 2006-2014 average for these inflows (29,078) is still significantly above 2015 figure. Again, I would suspect that much of this increase is accounted for by MNCs and also by demand for particular skills. Note: I will blogging on skills matters subsequently in a separate post.
  3. There has been virtually no change in inflows of people from the UK over the last 3 years, so nothing worth spotting here in terms of trends. Rest of EU-15 immigration flows also were relatively static, up to 8,900 in 2015 compared to 8,700 in 2014. Nonetheless, this has been the busiest year for EU15 migration inflows (ex-UK and Ireland) for some time - since 2009.
  4. Number of Irish nationals returning rose from 11,600 in 2014 to 12,100 in 2015. Which, kind of directly flies in the face of a number of media reports about 'returning migrants'. Apparently, the migrants are not quite returning, as current rate of immigration in Ireland by Irish nationals was the second slowest on record and much closer to the lowest year (2014) than to the third lowest (2013).


Now, consider emigration figures:


Again, few things worth a closer look:

  1. Irish emigration continued to decline in 2015 for the second year in a row. 2014 emigration of Irish nationals stood at 35,300 down from 40,700 in 2014 and down substantially on crisis period peak of 50,900 in 2013. The rate of emigration is now closer to 2006-2014 average of 29,444 that before, but it is still substantially above that number. Crucially, in more normal times, emigration by Irish nationals stood at around 13,700, which is well below current levels.
  2. Emigration by UK nationals out of Ireland remained pretty much stable and on-trend. Historical pre-2015 average is for annual outflow of 3,467 and in 2015 the number was 3,800. Emigration by the nationals of the EU15 states (ex-UK and Ireland) was up in 2015 at 15,600 compared to 14,000 in 2014. The rate is rising now for two years and is well ahead of 9,078 average for 2006-2014 period. This is interesting, as it reflects some shift in MNCs employment: in the past, MNCs were focusing much of their hiring on old EU markets, demanding language skills from these countries. Now, it seems the momentum is shifting toward ex-EU15 markets. Notably, pre-crisis average emigration by EU15 nationals stood at 6,667 per annum, very substantially below the 2015 figure.
  3. In contrast to EU15 pattern, emigration by the Accession EU12 nationals fell significantly in 2015 to 8,500 from 10,100 in 2014. This is the slowest rate of outflow for any year from 2007 on and significantly below the 2006-2014 average annual rate of outflow of 15,478.
  4. Rest of the World (ex-EU) emigration picked up, rising to 17,700 in 2015 compared to 14,100 in 2014 and against the 2006-2014 average of 9,9833. The reason for this, most likely, is the turnover of MNCs-employed tech workers and specialists who tend to stay in Ireland for 2-3 years and subsequently leave. 


So last remaking bit of analysis will have to cover net immigration / emigration:


As consistent with the number discussed above:

  1. Rate of net immigration from the 'Rest of the World' (ex-EU) picked up somewhat in 2015, rising to 12,700 from 11,200 in 2014. This is the highest rate of net increase in ex-EU population in Ireland for any year between 2006 and 2015.
  2. Second noticeable change in 2015 was positive contribution of EU12 (Accession states) nationals, with their net immigration at 4,300 in 2015 marking the first positive net result since 2008.
  3. EU15 (ex-UK and Ireland) net emigration remained significant and increased, with 6,700 more nationals of EU15 (ex-UK and Ireland) leaving Ireland than coming into Ireland in 2015, up on 5,300 in 2014. This marks the third year of rising net emigration by EU15 nationals out of Ireland and 6th consecutive year of negative net immigration by this group of residents.
  4. Irish nationals net emigration from Ireland remained very substantial in 2015 at 23,200. The number is lower than 29,200 net emigrations recorded in 2014 and the lowest reading in 4 years, but it is still well above the crisis period average. In simple terms, things are getting worse slower in this metric, they are not getting better.

Combined 2008-2015 net movements of people by nationality are shown in the chart below. Since 2008 through April 2015, there are 5,200 more UK nationals residing in Ireland, while the number of EU12 migrants rose 11,100. By far, the largest net emigration on a cumulative basis relates to outflow of Irish nationals: between 2008 and 2015, 132,400 more Irish nationals left the country than came back into the country - annual average rate of net emigration of 16,600 and in 2015 annual net emigration for Irish nationals was 6,700 above that.



29/8/15: Migration & Natural Changes in Irish Population: Top Level Analysis of 2015 data


Irish migration and population data for the 12 months through April 2015 have been published by CSO recently. It is a tough read. 

Let's take a look at overall picture (2015 here references May 2014-April 2015 period as per CSO data):
  • There were 67,000 births in 2015, down on 67,700 in 2014 - a decline of 700. Compared to peak births year - 2010 - births are down 10,200.
  • There were also 29,600 deaths in 2015, compared to 29,800 in 2014 which is 3,200 lower than peak year (1990), but more significantly - below 2013 and 2014 readings. 
  • Which means that natural increase in population was 37,400 in 2015, compared to 37,900 in 2014. This is the lowest natural rate of population increase since 2006 and it is driven exclusively by decline in birth rate which fell to 1.445% (births as percentage of total population) from 1.469% in 2014. Current birth rate is the lowest since 2001.
  • Immigration into Ireland amounted to 69,300 in 2015, an uplift on 60,300 in 2014 and the highest reading since 2009. Immigration has been increasing every year starting with 2013. Note: I will be blogging on quality of immigration and emigration separately, so stay tuned. Current rate of immigration is ahead of 2008-2014 average (64,500) but behind 2000-2007 average (80,100).
  • Meanwhile, emigration slipped slightly  - the good news you heard by now, most likely - from 81,900 in 2014 to 80,900 in 2015. This brings 2015 emigration closer to 2011 level (80,600) and lowest in 4 years. However, historical comparisons are still weak: in 2000-2007 average rate of emigration from Ireland was 30,700 and in 2008-2014 period it was 75,571, which means 2015 figure is higher than either pre-crisis average (obvious, really) and crisis period average. 
  • Net emigration stood at 11,600 in 2015, down from 21,400 in 2014 - a decline that is accounted for by an 8,700 increase in immigration and by 1,000 decrease in emigration. Thanks to immigrants numbers rising, we are at the lowest crisis period level of net emigration


Over 2009-2015, cumulated net outflows of people from Ireland stand at 153,800 - or 3.3% of our population in 2015 - close to 4 last years of combined natural increases in population (births less deaths). Over the last 6 years, average annual net emigration from Ireland stood at 25,900. During the 6 years of 1987-1992 the average was 21,050. even with 2015 decline in net emigration rate, we are still sending more people abroad in the current crisis than in the late-1980s -early 1990s crisis.

Which brings us to the 'opportunity cost' of this emigration, or in simple terms - what would our population be were it not for the crisis. 

First up, our current population estimates: per CSO, in April 2015 there were 4,635,400 people living in Ireland - an increase of 25,800 on same period 2014. The rate of increase was the highest since 2009. Which is all good news. However, the rate of annual population increase in 2015 was lower than 2008-2014 average (33,386) and was way below the 2000-2007 average of 80,100.

Now, let's take three scenarios. Starting with 1997 - the year when majority of us think Ireland's catching up with EA15 states was pretty much well underway and things were not yet fully 'mad' in a Celtic Garfield sense. This is also happens to be the year when net immigration posted the first above zero trend, with 7 year average through 1997 at 4,600 being the first positive 7 year average on record (note, 7 year average is chosen because of the cut-off period and because it corresponds to the duration of the current crisis too). Now, let's define 3 scenarios:
  1. Scenario 1: take an average for net migration over 1997-2000 (capturing the period all of can agree was pre-bubble in the property markets) and take projection from 1987 through 2015 at this average net immigration rate and accounting for actual realised natural rate of population change. You get 2015 population at 4,662,700 or 27,300 more people than current official estimated population.
  2. Scenario 2: take an average for net migration over 1997-2003 (7 years period), capturing the period that some think was still pretty much pre-Garfield craze and, crucially, before the Accession of 2004 that brought into Ireland significant inflows of Eastern European workers. You get 2015 population of 4,784,500 or 149,100 more people than current official estimated population.
  3. Scenario 3: while a bit outlandish, let's just consider the period of the entire Celtic Garfield and take the average net immigration rate at 1997-2007. Extreme, I know, but what the hell. You get 2015 population of 5,701,300 or 1,065,900 more people than current official estimated population.




Stay tuned for more analysis of net migration flows.

Friday, August 28, 2015

28/8/15: Core Retail Sales for July: Less of a Cheer, More of a Smile


With much of hullabaloo around it, the Retail Sales figures for July were published today. The CSO headline on the matter read: "Retail Sales Volume increased by 11.6% in July 2015". Which is, of course, correct... to a point. The figure references sales inclusive of automotive sales. And it references volumes of sales.

So here are the actual retail sales figures, for retail sales excluding motors.

First, consider seasonally-adjusted sales allowing m/m comparatives:

  • Value of core (ex-motors) retail sales increased 0.3% m/m in July and this only partially (albeit substantially) corrects for 0.3% decline m/m in June. Compared to 2005 average level, value of sales today is only 1.09% higher, which is... before inflation is factored in. 3mo MA of Value indices in July was down 0.03% on 3mo MA average through June, while a month ago the same was 0.7% higher. In other words, there is nothing 'convincing' in the value of sales data. And this is concerning, because retailers don't get their revenues and profits from volumes of sales. They get them from value of sales.
  • Volume of retail sales (ex-motors) was up 0.64% m/m in July, having previously posted a decline of 0.12% in June. So July volumes of sales significantly over-compensated for June decline. Which is good news. Compared to 2005 average, July figure is 12.1% higher for the volume of sales, which means that deflation has resulted in more sales by volume, but barely any change in value: selling more stuff but getting less per unit sold is the retailers' margin nightmare and it has been going on for some years now. But the good news on Volume run out when you consider 3mo MA: 3mo MA through July was down 0.12% compared to 3mo MA through June, having previously been up 0.7%. So on 3mo MA basis (smoothing a bit volatility) value and volume of retail sales both fell in July.
  • Meanwhile, the never-ending exuberance of Irish consumers, as measured by Consumer Confidence Index posted some moderation in July, falling 3% m/m. Still Consumer Confidence in July 2015 is 97% (that's right - 97%) higher than the 2005 average. You really gotta wonder...
Two charts to illustrate the above trends:


As can be seen from the chart above, there is now divergence in the series for Value (rising slower) and Volume (rising faster) of core retail sales. This, with Value of sales running now persistently above the trend (suggesting risk of downward correction in the future), whilst Volume series running along the trend. Volume is converging with Consumer Confidence, while Value is diverging. Closer look at the latter next:


 And now to y/y data based on unadjusted series:



Per data charted above y/y changes were:

  • Value of core retail sales rose 3.40% y/y - which is a good performance, but not exactly stellar. In June, y/y increase was 2.0% and in July 2014 y/y rise was 1.2% which means this July growth was stronger. However, pre-crisis average y/y growth rate in Value of retail sales was 6.93% and this means that current rate of increase is just under 1/2 the rate of average rise in pre-crisis years. Smoothing out some volatility, 3mo average through July was 99.3 which is stronger than 3mo average through April 2015 (93.6) and is up 3% on 3mo average through July 2014. These are good figures, no arguing there.
  • Volume of retail sales, predictably, was up 6.7% y/y in July - double the rate of growth in Value of sales and above the pre-crisis average rate of growth of 6.2%. Volume of sales gained in July in annual rate of growth compared to June (4.7%) and compared to July 2014 (3.2%). And on 3mo average basis, the index was up 6% y/y for 3mo through July - double the rate of growth in Value.
  • Hence, overall we have the same picture in unadjusted data: rates of growth in Value of sales are healthy, but not spectacular, while rates of growth in Volume are strong. Volume is diverging from Value and there is nothing new here - it has been thus since the end of 2013.

Good news is that on 3mo average basis, May-July 2015 figures were

  • Positive in y/y terms for majority sub-sectors in value terms (excluding Food, Beverages & Tobacco and Motor Fuel) and for all sub-sectors in volume terms
  • The picture was a bit more fragmented for 3mo change through July compared to 3mo change through April, as shown in the table below.

Thus, overall, there are some good news in the retail sales figures. Do they warrant a huge wave of congratulatory backslapping exercises in the media? No. Do they warrant much of optimism that the sector is experiencing a big revival? Not exactly. 



28/8/15: Gold & Silver: Does technical analysis beat the market?


An interesting piece of research co-authored by Brian Lucey on efficacy of technical analysis in gold & silver markets: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2652637.

"This paper studies whether intraday technical trading rules produce significant payoffs in the gold and silver market using three popular moving average rules."

And the conclusions are (emphasis is mine): "The initial results show that the SMA, WMA and EMA trading rules generate significant negative payoffs using the parameters common in the literature in the high-frequency gold and silver markets. This suggests that there is no significant profit to be gained from technical trading in the gold and silver markets. However, our parameter sweep results show that there are a number of parameter combinations that generate significant profit in the gold market, but none in the silver market. Further, the best performing rules have different parameters to those used the existing literature. We show that longer run averages should be used by investors on intraday data and that investors need to employ different parameters when utilising technical analysis on daily and intraday data. In order to examine whether investors could have actually utilised the best performing rules, we perform an in- and out-of-sample test and show that only the SMA rule for gold generates significant profits in the in-sample as well as the out-of-sample period. All of the other best rules in the in-sample period generate either insignificant or negative payoffs in the out-of-sample period."

28/8/15: Inflation Expectations: Euro and U.S.


Having earlier posted a chart on Central Banks balancesheets expansion (see here), here is an interesting chart plotting inflation expectations (5yr5yr swaps - effectively markets expectations for 5 years from now inflation average over subsequent 5 years)


The above shows that although there has been an uplift in Euro area inflation expectations over the course of 2015 to-date, consistent with QE carried out by the ECB, the expectations have tanked since the start of Q3 2015 in line with those in the U.S.

More ominously, expectations remain in the territory where neither the Fed nor the ECB are capable of convincingly exiting monetary easing.

While the U.S. expectations are closer to target (at 2.23%) but still weak, Euro area expectations are exceptionally weak at 1.63%. Gotta do some more printing (for ECB) and less talking about tapering (for both the Fed and the ECB)...

28/8/15: Patents & R&D expenditure effects on equity returns in pharma industry


Martina Feyzrakhmanova and myself have a new paper out with Applied Economics Letters titled "Patents and R&D expenditure effects on equity returns in pharmaceutical industry". Working paper (ungated version) is available here: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2438461.

28/8/15: CBR and Ruble: Fiscal Balance in Oil's Shadow


As I noted earlier this month, Russia has officially entered the recession. The key drivers for 3.4% contraction in 1H 2015 were the same as the key pressures on growth back in 2H 2014: oil prices, investment collapse on foot of high interest rates, inflationary environment that restricts CBR's room for cutting rates, and sanctions (or rather geopolitical risks and pressures, linked in part to sanctions).

That said, in June and early July there were some hopes for economy starting to stabilise, although fixed investment was down 7.1% y/y in June, marking 18th consecutive month of y/y declines. These are now once again under pressure and the cause is... oil price.

Here is how closely paired has Russian Ruble been to oil prices in trend terms since July 2014, although correlation was weaker in preceding period. Overall, as the chat shows, there are two very distinct periods of Ruble valuations catch up with oil prices: June 2014-February 2015 and mid-July 2015 through present.


Russia's Central Bank is switching between little concern for Ruble to interventions and back to staying out of the markets appears to be more than a simply random walk. Instead, it is a game consistent with rebalancing Ruble valuations to fit budgetary dynamics.

The reason for this is that (as shown in the chart above) Ruble strengthening above oil-linked fundamentals earlier this year was an actual threat to budgetary dynamics, and over the last couple of weeks, correcting valuations of Ruble re-established closer connection to oil prices. Hence, in July, CBR managed to deliver a shallower cut to interest rates (-50bps) compared to June (-100bps).

With CBR continuing to stick to its June 2016 forecast for inflation to fall to 'under 7%' by then and hit 4% in 2017 compared to July 2015 CPI at 15.6%, Russia went on to issue its first CPI-linked bonds / linkers amounting to RUB75 billion (OFZ-IN, 8 year notes) at 91% of nominal, on cover of RUB200 billion (more than 25% of demand coming from foreign investors). Real yield at issuance was 3.84% - relatively high-ish, implying underpricing of the bond in a market with relatively hefty demand and forward expectations for significant easing in inflation. Something is slightly amiss here.

In line with up-down interventions, the CBR continued to trend flat on foreign exchange reserves. End of June, total Russian FX reserves stood at USD361.575 billion, and by end of July these fell to USD357.626 billion. As of last week, the reserves were back up at USD364.6 billion.


Weekly data from CBR does not allow for compositional analysis of reserves, but looking at the monthly data the pattern repeats.


Actual liquid FX reserves and gold stood at USD347.1 billion at the end of July against USD350.957 billion at the end of June. This is barely up on end-April period low of USD345.373 billion, although well within the FX- and gold-valuations range of change.

Meanwhile, data through July 2015 shows net purchases of dollars of USD3.76 billion against USD3.831 billion in June and USD2.531 billion in May by CBR. Overall, from January 2015 through July 2015, CBR bought (net) USD7.8 billion and there were no net purchases/sales of euro.


All of the above suggests that CBR will likely resume rate cuts if Ruble firms up from its recent valuations. Two weeks ago, RUB/USD was at 64.947 (72.197 to Euro), peaking at 70.887 four days ago (82.373 to the Euro) and currently at 66.8875 (74.984 to Euro), not exactly warranting a move by the CBR yet, but back in the relative comfort zone for the Bank to sit on its hands once again.