Wednesday, August 16, 2017

16/8/17: Year Eight of the Great American Recovery: Household Debt

U.S. data for household debt for 2Q 2017 is out at last, and the likes of Reuters and there best of the official business media are shouting over each other about the ‘record debt levels’ warnings. As if the ‘record debt levels’ is something so refreshingly new, that no one noticed them in 1Q 2017.

So with that much hoopla in your favourite media pages, what’s the data really telling us?

Quite a bit, folks. Quite a bit.

Let’s start from the top:

Debt levels are up. Almost +4.5% y/y. All debt categories are up, save for HE Revolving debt (down 5.44% y/y). Increases are led by Auto Loans (+7.89% y/y) and Credit Cards (+7.54%). High growth is also in Student Loans (+6.75%). Mortgages debt is rising much slower, as consistent with lack of purchasing power amongst the younger generation of buyers.

As you know, I look at this debt from another perspective, slightly different from the rest of the media pack. That is, I am interested in what is happening with assets-backed debt and asset-free debt. So here it is:

Yes, debt is up again. Mortgages debt share of total household debt has shrunk (it is now at 67.7%) and unsecured debt share is up (32.3%). Unsecured debt was $3.925 trillion in 2016 Q2 and it is now $4.148 trillion. Why this matters? Because although cars can be repossessed and student loans are non-defaultable even in bankruptcy, in reality, good luck collecting many quarters on that debt. Housing debt is different, because with recent lending being a little less mad than in 2004-2007, there is more equity in the system so repossessions can at least recover meaningful amounts of loans. So here’s the thing: low recovery debt is booming. While mortgages debt is still some $600 billion odd below the pre-crisis peak levels.

On the surface, mortgages originations are improving in terms of credit scores. In practice, of course, credit scores are superficially being inflated by all the debt being taken out. Yes, that’s the perverse nature of the American credit ratings system: if you have zero debt, your credit rating is shit, if you are drowning in debt, you are rocking…

Still, here is the kicker: mortgages credit ratings at origination are getting slightly stronger. Total debt written to those with a credit score <660 2016.="" 2016="" 2017="" 2q.="" 2q="" also="" auto="" billion="" buyers="" class="Apple-converted-space" credit="" down="" fell="" from="" good="" improving:="" in="" is="" issuance="" loans="" news.="" origination="" quality="" score="" span="" sub-660="" to="" which=""> 

Bad news:

Severely Derogatory and 120+ delinquent loans are still accounting for 3% of total loans, same as in 2Q 2016 and well above the pre-crisis average of 2.1%. Total share of delinquent loans is at 4.77%, slightly below 1Q 2017 (4.83%) and on par with 4.79% a year ago. So little change in delinquencies as a result of improving credit standards at origination, thus. Which suggests that improving standards are at least in part… err… superficial.

And things are not getting better across majority of categories of delinquent loans:

As the above clearly shows, transition from lesser delinquency to serious delinquency is up for Credit Cards, Student Loans and Auto Loans. And confirming that the problem of reading Credit Scores as improvement in quality of borrowers are the figures for foreclosures and bankruptcies. These stood at 308,840 households in 2Q 2017, up on 294,100 in 1Q 2017 and on 307,260 in 2Q 2016. Now, give it a thought: over the crisis period, many new mortgages issued went to households with better credit ratings, against properties with lower prices that appreciated since issuance, and under the covenants involving lower LTVs. In other words, we should not be seeing rising foreclosures, because voluntary sales should have been more sufficient to cover the outstanding amounts on loans. And that would be especially true, were credit quality of borrowing households improving. In other words, how does one get better credit scores of the borrowers, rising property prices, stricter lending controls AND simultaneously rising foreclosures?

Reinforcing this is the data on third party debt collections: in 2Q 2017, 12.5% of all consumers had outstanding debt collection action against them, virtually flat on 2Q 2016 figure of 12.6%. 

In simple terms, in this Great Recovery Year Eight, one in eight Americans are so far into debt, they are getting debt collectors visits and phone calls. And as a proportion of consumers facing debt collection action stagnates, their cumulative debts subject to collection are rising. 

Things are really going MAGA all around American households, just in time for the Fed to hike cost of credit (and thus tank credit affordability) some more. 

Tuesday, August 15, 2017

15/8/17: A Great Recovery or a Great Stagnation?

Value-added is one measure of economic activity that links the production side to consumption/ demand side (using inputs of say $X value to produce a good that sells for $Y generates $Y-$X in Gross Value Added). Adjusted for inflation, this returns Real Gross Value Added (RGVA) in the economy. Taken across two key sectors that comprise the private sector economy: households & institutions serving the households, and private businesses (including or excluding farming sector), these provide a measure of the economic activity in the private economy (i.e. excluding Government).

Since the end of WW2, negative q/q growth rates in the private sectors RGVA have pretty accurately tracked evolution of economic growth (as measured, usually, by growth rates in GDP). Only in the mid-1950s did the private sector RGVA growth turn negative without triggering associated official recession on two occasions, and even then the negative growth rates signalled upcoming late-1950s recession.

Which brings us to the current period of Great Recovery.

Consider the chart below, computed based on the data from the Fred database:

The first thing that jumps out in the above data is that since the end of the Great Recession, the period of the Great Recovery has been associated with two episodes of sub-zero growth in the private sector RGVA. This is unprecedented for any period of recovery post-recession, except for the period between two closely-spaced 1950s recessions: July 1953-April 1954 and August 1957-March 1958.

The second thing that stands out in the data is the average growth rate in RGVA during the current recovery. At 0.579% q/q, this rate is the lowest on the record for any recovery period since the end of WW2. Worse, it is not statistically within 95% confidence interval bands for average growth rate in post-recovery periods for the entire history of the U.S. economy between January 1948 and October 2007. In other words, the Great Recovery is, statistically, not a recovery at all.

The third matter worth noting is that current non-recovery Great Recovery period is the third consecutive period of post-recession growth with declining average growth rates.

The fourth point that becomes apparent when looking at the data is that the current Great Recovery produced only two quarters with RGVA growth statistically above the average rate of growth for a 'normal' or average recovery. This is another historical record low (on per-annum-of-recovery basis) when compared across all other periods of economic recoveries.

All of the above observations combine to define one really dire aftermath of the Great Recession: despite all the talk about the Great Recovery sloshing around, the U.S. economy has never recovered from the crisis of 2007-2009. Omitting the years of the official recession from the data, the chart below shows two trends in the RGVA for the private sector economy in the U.S.

Based on quadratic trends for January 1948-June 2007 (pre-crisis trend) and for July 2009 - present (post-crisis trend), current recovery period growth is not sufficient to return the U.S. to its pre-crisis long term trend path. This is yet another historical first produced by the data. And worse, looking at the slopes of the two trend lines, the current recovery is failing to catch up with pre-crisis trend not because of the sharp decline in real economic activity during the peak recession years, but because the rate of growth post-Great Recession has been so anaemic. In other words, the current trend is drawing real value added in the U.S. economy further away from the pre-crisis trend.

The Great Recovery, folks, is really a Great (near) Stagnation.

Sunday, August 13, 2017

12/8/17: Some growth optimism from the Russian regional data

An interesting note on the latest data updates for the Russian economy via Bofit.

Per Bofit: "Industrial output in Russian regions rises, while consumption gradually recovers." This is important, because regional recovery has been quite spotty and overall economic recovery has been dominated by a handful of regions and bigger urban centres.

"Industrial output growth continued in the first half of this year in all of Russia’s eight federal districts," with production up 1.5–2% y/y in the Northwest, Central and Volga Federal Districts, as well as in the Moscow city and region. St. Petersburg regional output rose 3-4% y/y.

An interesting observation is that during the recent recession, there has been no contraction in manufacturing and industrial output. Per Bofit: "Over the past couple of years, neither industrial output overall nor manufacturing overall has not contracted in any of Russia’s federal districts. Industrial output has even increased briskly in 2015–16 and this year in the Southern Federal
District due to high growth in manufacturing and in the Far East Federal District driven by growth in the mineral extraction industries."

This is striking, until you consider the nature of the 2014-2016 crisis: a negative shock of collapsing oil and raw materials prices was mitigated by rapid devaluation of the ruble. This cushioned domestic production costs and shifted more demand into imports substitutes. While investment drop off was sharp and negative on demand side for industrial equipment and machinery, it was offset by cost mitigation and improved price competitiveness in the domestic and exports markets.

Another aspect of this week's report is that Russian retail sales continue to slowly inch upward. Retail sales have been lagging industrial production during the first 12 months of the recovery. This is a latent factor that still offers significant upside to future growth in the later stages of the recovery, with investment lagging behind consumer demand.

Now, "retail sales have turned to growth, albeit slowly, in six [out of eight] federal districts."

Here is why these news matter. As I noted above, the recovery in Russian economy has three phases (coincident with three key areas of potential economic activity): industrial production, consumption and investment. The first stage - the industrial production growth stage - is on-going at a moderate pace. The 0.4-0.6 percent annual growth rate contribution to GDP from industrial production and manufacturing can be sustained without a major boom in investment. The second stage - delayed due to ruble devaluation taking a bite from the household real incomes - is just starting. This can add 0.5-1 percent in annual growth, implying that second stage of recovery can see growth of around 2 percent per annum. The next stage of recovery will involve investment re-start (and this requires first and foremost Central Bank support). Investment re-start can add another 0.2-0.3 percentage points to industrial production and a whole 1 percent or so to GDP growth on its own. Which means that with a shift toward monetary accommodation and some moderate reforms and incentives, Russian economy's growth potential should be closer to 3.3 percent per annum once the third stage of recovery kicks in and assuming the other two stages continue running at sustainable capacity levels.

However, until that happens, the economy will be stuck at around the rates of growth below 2 percent.

Saturday, August 12, 2017

12/8/17: Are Irish Property Prices on a Sustainable Path?

Some of the readers of this blog have been asking me to revisit (as I used to do more regularly in the past) the analysis of Irish property prices in relation to the ‘sustainability trend’. With updated CSO data on RPPI, here is the outrun.

The charts below show current National and Dublin property price indices in relation to the trends computed on the basis of the following CORE assumptions:
  • Starting period: January 2005
  • Starting index ‘sustainability’ positions: National = 82.0 (implying that long-term sustainable market valuations were around 18 percentage points below market price levels at January 2005 or at the levels comparable to Q4 2010); Dublin = 83.0 (implying 17 percentage points discount on January 2005).

Charts above use the following SPECIFIC trend assumptions:
  • Linear (simplistic) trend at 2% inflation target + 0.5 percentage points margin. This implies that under this trend, property prices should have evolved broadly-speaking at inflation, plus small margin (close to tracker mortgage rate margin).

In all cases, current markets valuations are well below the long-term sustainability target and there is significant room for further appreciation relative to these trends (see details of target under-shooting in the summary table below).

Chart above shows tow series sustainability targets computed on the basis of different specific assumptions, while retaining same core assumptions:
  • I assume that property prices should be sustainably anchored to weekly earnings. 

Using only weekly earnings evolution over January 2005-present, as shown in the above chart, both Dublin and National house prices are currently statistically at the levels matching sustainability criteria. There is no statistical overshooting of the sustainability bounds, yet.

Chart above again modifies specific assumptions, while retaining the same core assumptions. Specifically:
  • I assume that both earnings and interest rates (using Euribor 12 months rate as a dynamic gauge) co-determine sustainable house prices. In a away, this allows us to reflect on both income and cost of debt drivers for house prices.

As the chart above clearly shows, both National and Dublin property markets are still well underpriced compared to the long term sustainability targets, defined based on a combination of earnings and interest rates. Note: correcting this chart for evolution of unemployment brings sustainability benchmarks roughly half of the way closer to current prices, but does not fully erase the gap.

Summary table below:

So, overall, the above exercise - imperfect as it may be - suggests no evidence of excessive pricing in Irish residential property at this point in time. There are many caveats that apply, of course. Some important ones: I do not account for higher taxes; and I do not factor in difficulties in obtaining mortgages. These are material, but I am not sure they are material enough to bring the above gaps to zero or to trigger overpricing. Most likely, the national residential prices are somewhere around 5-7 percent below their sustainability bounds, while Dublin prices are around 7-10 percent below these bounds. Which means we have a short window of time to bring the markets to the sustainable price dynamics path by dramatically altering supply dynamics in the property sector. A window of 12-18 months, by my estimates.

Thursday, August 10, 2017

10/8/17: 2Q Start Ups Funding Data: Big Lessons Coming

2Q 2017 figures for seed funding and startups capital rounds is in, and the slow bleeding of the Silicon Dreams appears to be entering a new, accelerated stage. 

Seed funding posted continuous declines over the last two years, falling some 40 percent on 3Q 2015 peak in terms of number of transactions and down 24 percent in volume. 

Source: PitchBook Inc

Combined seed and angel investment deals numbered just 900 on 2Q 2017, down 200 on same period in 2016 and well below ca 1,500 deals completed in 2Q 2015. In volume terms, 2Q 2017 came in with total investment of $1.65 billion, down on $1.75 billion in 2Q 2016 and $2.19 billion in 2Q 2015. Masking these falls somewhat, terms of investments have tightened significantly over the last two years, meaning that actual in-hand capital allocations have fallen more than the headline volume figures suggest.

There are some reasons for this decline. Firstly, recent tech IPOs signal Wall Street’s growing scepticism over unicorns valuations of tech companies. Secondly, the quality of new deals coming into the market is slipping: if two years ago everyone was chasing mushrooming sector of ‘Uber-for-X’ companies, today the ‘disruption’ pitch is getting old and the focus might be shifting on later stage financing of already existent companies.Thirdly, investment funds are facing internal problems - the classical allocation dilemmas. As funds under management rise in the angel and VC investment outfits, it becomes harder and harder for them to meaningfully allocate small investments to smaller start ups. They become more dependent on ‘finding the next Uber’. As a result, the funds are shifting their cash to already existent early stage companies, away from angel and seed finance. 

To see the latter two points, consider the median seed deal size. Two years ago, that stood at around $500,000. Today, it is at around $1.5 million. 

There is also the issue of timing. Boom in seed funding in tech sector is now good decade long. And it is time to count the proverbial chickens. As the industry pursued the investment model of ’spray the cash around and pray for a return somewhere’, a range of seed finance funds are closing down and posting poor returns. This, in turn, makes new investors more cautious.

Why this is significant? Because many tech start ups generate no meaningful revenues and, even at later stages of development, once revenues ramp up, they tend to run huge losses. The reason behind this is that many tech start ups (especially the larger ones) pursue business models based on aggressive expansion of market share in markets (e.g. taxis and food deliveries) where traditional business margins are already thin. In other words, by pursuing volume, not profit, the start ups must use increasing injections of capital and large capital allocations up front to stay afloat. 

This, of course, is not a sustainable model for business development. But tell that to the politicians and business leaders and investors, all of whom tend to chase size before understanding that business needs to make profits before it can raise employment and build brand dominance.

So for the future, folks: stop chasing pre-revenue, business plan (or tech platform)-based funding. Focus on generating your first sales and showing these to have margin potential. Remember, corporate finance matters not so much on the capital budgeting side, but on the cash flow spreadsheet. 

Wednesday, August 9, 2017

9/8/17: Global Debt Bubble 2002-2017

A nice chart showing evolution of global debt levels since 2002 via Reuters:


9/8/17: Euro Area Banks Bailouts: The Legacy Still Hangs Over Our Heads

The Financial Times has published a very neat visualisation of the global banks bailouts net impact to-date:

And the snapshot magnifying European states impact:

None of the Euro area states have recovered all funds deployed in bailing out the banks. And the worst performer of all states is Ireland.

Note: the chart references bailouts as a share of GDP. Of course, in the case of Ireland and Cyprus, GDP is by a mile (in the case of Ireland, by about one third) is unrepresentative of the actual national income available to sustain these.

Another note: the three worst-hit countries, Ireland, Greece and Cyprus, all remain deeply under water when it comes to recovering funds spent in the bailouts, even though the three had, on the surface, different bailout regimes applied. Specifically, Cyprus (and to a lesser extent, Greece) was supposed to be a model bailout, serving as the basis for the future bailouts across the Euro area (including structured bail-ins of private depositors).

So much for the hope of the Euro area 'reforms' working... And so much for the end of the Crisis...

8/8/17: Did Irish Household Spending Fully Recover from the Crisis?

I have recently seen several research notes claiming that in 1Q 2017, Ireland has finally fully recovered from the shock of the Great Recession. These claims were based on consumer demand regaining its pre-crisis peak.

What do the facts tell us about this claim? That it is a half-truth.

Consider the following chart plotting consumer demand (consumer expenditure on goods and services) computed on an aggregate 4 quarters running basis. I use official CSO data for both expenditure figures and population figures. And I compute per-capita expenditure on the basis of these statistics.

In 1Q 2017, aggregate household expenditure on goods and services stood at EUR96.16 billion against pre-crisis peak of EUR94.118, using constant prices to account for official inflation. Incidentally, there is nothing new in the claim of recovery on that basis, because Irish households' aggregate spending on goods and services has surpassed pre-crisis peak in 2Q 2016.

The problem with the aggregate expenditure figure is that population changes. So the chart above also shows per-capita real expenditure, expressed in 1,000s of constant euros. Here, the matters are a bit less impressive. Per capita household expenditure on goods and services in Ireland peaked pre-crisis at EUR21,508.75. At the end of 1Q 2017, this figure was EUR 20,574.71.

There is another problem with analysts' celebrations of the 'end of the lost decade'. Aggregate household expenditure peaked (pre-crisis) in 1Q 2008, so it took 32 quarters to recover that peak. Per-capita household expenditure peaked in 2Q 2008, which means we are 35 quarters into the crisis and counting. Neither comes up to a full decade.

Finally, there is a really big problem. This one relates to what a 'recovery from the crisis' really means. In the above, we implicitly assume that a recovery from the crisis is return to pre-crisis peak. But there is a major problem with that, because our current state of life-cycle incomes, savings and debt in part reflect decisions made under the assumptions that operated back in the pre-crisis period. In other words, our income, savings, investment, career choice and debt carry a 'memory' of the times when (pre-crisis) trends did not incorporate any expectation of the crisis.

What does this mean? It means that psychologically, materially and even economically, the end of the crisis is when the economy returns to where it should have been were the pre-crisis trend extended into the present. To make this comparative more robust, we should also recognise that, in part, the pre-crisis trend should have omitted at least some of the most egregious excesses of the bubble years.

Let's do that exercise, then. Let's take pre-crisis trend in household expenditure (aggregate and per-capita) for year 3Q 2000-2004 (eliminating the explosive years of 1997-2000 and 2005-2007) and see where we are today, compared to that trend.

On trend, our aggregate personal expenditure should have been around EUR111.7 billion marker in 1Q 2017. It was EUR96.16 billion. This hardly reflects a recovery to the pre-crisis trend.

Also on trend, our per capita expenditure should have been around EUR24,140 in 1Q 2017. It was EUR20,575. This hardly reflects a recovery to the pre-crisis trend.

As some of my friends in Irish stuffbrokerages have been known to remark in private: "Shit! Damn numbers." Indeed... the recovery will have to wait... but, lads, you know you can do these calculations yourselves, right? You are paid six figure salaries and bonuses to do them. Or may be you are not. May be, you are paid six figure salaries and bonuses not to do these calculations...

Tuesday, August 8, 2017

8/8/17: Irish Taxpayers Face a New Nama Bill

Ireland has spent tens of billions to prop up schemes, like Nama and IBRC. These organisations pursued developers with a sole purpose: to bring them down, irrespective of the optimal return strategy from the taxpayers perspective and regardless of optimal recovery strategies for asset recovery. We know as much because we have plenty of evidence - that runs contrary to Nama and IBRC relentless push for secrecy on their assets sales - that value has been destroyed during their workout and asset sales phases. We know as much, because leaders of Nama have gone on the record claiming that developers are, effectively speculators, 'good for nothing else, but attending Galway races', and add no value to construction projects.

Now, having demolished experienced developers and their professional teams, having dumped land and development sites into the hands of vulture investors, who have no expertise nor incentives to develop these sites, the State has unrolled a massive subsidy scheme to aid vultures in developing the sites they bought on the State-sponsored firesales.

As an aside, this June, Nama officially acknowledged the fact that majority of its sales of land resulted in no subsequent development. What Nama did not say is that the 'developers' hoarding land are the vulture funds that bought that land from Nama, just as Nama continued to insist that its operations are helping the construction and development markets.

Why? Because Nama was set up with an explicit mandate to 'help the economy recover' and to drive 'markets to restart functioning again', and to aid social housing crisis (remember when in 2012 - five years ago - Nama decided to 'get serious' about social housing?). And Nama has achieved its objectives so spectacularly, Ireland is now in the grips of a housing crisis, a rental market crisis and a cost-of-living crisis.

Read and weep:
Irish taxpayers are now paying the third round of costs of the very same crisis: first round of payments went to Nama et al, second to the banks, and now to the 'developers' who were hand-picked by Nama and IBRC to do the job they failed to do, for which Nama was created in the first place.

Oh, and because you will ask me when the fourth round of payments by taxpayers will come due, why, it is already in works. That round of payments will cover emergency housing provision for people bankrupted by the banks and Nama-supported vultures. That too is on taxpayers shoulders, folks...

Friday, August 4, 2017

3/8/17: New research: the Great Recession is still with us

Here is the most important chart I have seen in some months now. The chart shows the 'new normal' post-2007 crisis in terms of per capita real GDP for the U.S.


The key matters highlighted by this chart are:

  1. The Great Recession was unprecedented in terms of severity of its impact and duration of that impact for any period since 1947.
  2. The Great Recession is the only period in the U.S. modern history when the long term (trend) path of real GDP per capita shifted permanently below historical trend/
  3. The Great Recession is the only period in the U.S. modern history when the long term trend growth in GDP per capita substantially and permanently fell below historical trend.
As the result, as the Roosevelt Institute research note states, " output remains a full 15 percent below the pre-2007 trend line, a gap that is getting wider, not narrower, over time".

The dramatic nature of the current output trend (post-2007) departure from the past historical trend is highlighted by the fact that pre-crisis models for forecasting growth have produced massive misses compared to actual outrun and that over time, as new trend establishes more firmly in the data, the models are slowly catching up with the reality:

Source: ibid

Finally, confirming the thesis of secular stagnation (supply side), the research note presents evidence on structural decline in labor productivity growth, alongside the evidence that this decline is inconsistent with pre-2007 trends:

On the net, the effects of the Great Recession in terms of potential output, actual output growth trends, labor productivity and wages appear to be permanent in nature. In other words, the New Normal of post-2007 'recovery' implies permanently lower output and income. 

3/8/17: BRIC Composite PMIs: July

Having covered BRIC Manufacturing PMIs in the previous post (, and Services PMIs (, here is the analysis of the Composite PMIs.

Table below summaries current shorter term (monthly) trends in Composite PMIs:

Brazil has slipped into a new sub-50 Composite PMI trend in 2Q 2017 and, as of July, remains in the slump, although at 49.4, July Composite PMI reading signals much weaker rate of economic activity contraction than the June reading of 48.5. The problem for Latin America’s largest economy is that the hopes for an extremely weak recovery, set in 50.4 readings in April and May are now gone. In fact, 2Q 2017 average Composite PMI for Brazil stood at 49.8, which was stronger than July reading and marked the strongest performance for the economy since 3Q 2014. All in, July marked the start of the 14th consecutive quarter of Composite PMIs signalling economic recession.

Russia Composite PMI at the end of July stood at 53.4, a respectably strong number, signalling good growth prospects for the economy, but down from 54.8 in June and 56.0 in May. In fact, July reading was the lowest in 9 months. Given the economy’s performance in 1Q 2017, set against composite PMIs, the July and 1-2Q readings suggest that Russia is on track to record 1.0-1.5% growth this year, but not quite 2.0% or higher as expected by the Government. We will need to see 3Q and 4Q averages closer to 56-57 range to have a shot at above 1.5% growth.

China posted 2Q 2017 Composite PMI at 51.3, which is below July 51.9 reading. Still, July improvement is yet to be confirmed across the rest of 3Q 2017. China’s Composite PMI slowed from a recent peak of 53.1 in 4Q 2016 to 42.3 in  1Q 2017 and 51.3 in 2Q 2017.

India’s Composite PMI reflected wide-ranging weakening in the economy struck by both botched de-monetisation ‘reform’ and equally bizarre tax reforms. Sinking from appreciably strong 52.2 in 2Q 2017 to 46.0 in July, this fall marked the lowest PMI reading since 1Q 2009 and the second lowest reading on record. India’s economy has been in a weak state since 3Q 2016 when Composite PMI averaged 53.1. The PMI fell to 50.7 and 50.8 in 4Q 2016 and 1Q 2017 before recovering in 2Q 2017. This recovery is now in severe doubt. We will need to see August and September readings to confirm an outright PMI recession, but the signs from July reading are quite poor.

All in, in July, Russia was the only BRIC economy that came close (at 53.4) to Global Composite PMI reading of 53.5. Two BRIC economies posted a sub-50 reading. In 2Q 2017, Global Composite PMI was 53.7, with Russia Composite PMI at 55.4 being the only BRIC economy that supported global economic growth to the upside. In fact, Russia lead Global Composite PMIs in every quarter since  2Q 2016.

3/8/17: The tale of Irish tax receipts tails

My article for the Sunday Business Post  on the latest trend in the Irish Exchequer receipts and how it stacks up against Budget 2018 prospects for tax cuts. Link here:

Thursday, August 3, 2017

3/8/17: BRIC Services PMI: July

Having covered BRIC Manufacturing PMIs in the previous post (, here is the analysis of the Services Sector PMIs.

Brazil Services PMI continued trending below 50.0 mark for the third month in a row, hitting 48.8 in July, after reaching 47.4 in June. While the rate of contraction in the sector slowed down, it remains statistically significant. This puts an end to the hope for a recovery in the sector, with Brazil Services PMIs now posting only two above-50 (nominal, one statistically) readings since October 2014.

Russian Services PMI also moderated in July, although the reading remains statistically above 50.0. July reading of 52.6 signals slower growth than 55.5 reading in June. The Services sector PMIs are now 18 months above 50.0 marker, continuing to confirm relatively sustained and robust (compared to Manufacturing sector) expansion.

China Services PMI remained in the statistical doldrums, posting 51.5 in July gayer 51.6 in June. The indicator has never reached below 50.0 in nominal terms in its history, so 51.5 reading is statistically not significant, given PMIs volatility and positive skew. Overall, this is second consecutive month of PMIs falling below statical significance marker, implying ongoing weakness in the Services economy in China.

India’s Services PMIs followed Manufacturing sector indicator and tanked in July, hitting 45.9 (sharp contraction), having previous posted statistically significant reading for expansion at 53.1 in June. Volatility in India’s Services indicator is striking.

Table and chart below summarise short term movements:

Looking at quarterly comparatives, July was a poor month for Brazil Services sector, with July reading of 48.8 coming in weaker than already poor 49.0 indicator for 2Q 2017. In Brazil’s case, current recession in Services is now reaching into 12th consecutive quarter in nominal terms and into 15ht consecutive quarter in statistical terms. Russia Services PMI also moderated at the start of 3Q 2017 (52.6 in July) having posted average 2Q 2017 PMI of 56.0. Russia Services sector expansion is now into its 6th consecutive quarter (statistically) and seventh consecutive quarter nominally. The same, albeit less pronounced, trend is also evident in China (July PMI at 51.5 against 2Q 2017 PMI of 52.0). India Services PMI was under water in 4Q 2016, followed by weak (zero statistically) growth in 1Q 2017 and somewhat stronger growth in 2Q 2017. The start of 3Q 2017 has been marked by a sharp, statistically significant negative growth signal.

With Global Services PMI hitting 53.7 in July, against 53.8 average for 2Q 2017 and 53.6 average in 1Q 2017, BRIC economies overall are severely underperforming global growth conditions (BRIC Services PMI is now below Global Services PMI in 3 quarters running and this trend is confirmed at the start of 3Q 2017).

3/8/17: BRIC Manufacturing PMIs: July

BRIC PMIs for July 2017 are out, so here are the headline numbers and some analysis. 

Top level summary of monthly readings for BRIC Manufacturing PMIs is provided in the Table below:

Of interest here are:
  • Changes in Brazil Manufacturing PMI signalled weakening in the economy in June that was sustained into July. Manufacturing PMI for Brazil has now fallen from 52.0 in May to 50.5 in June and to 50.0 in July. This suggests that any recovery momentum was short lived. 
  • Russian Manufacturing PMI, meanwhile, powered up to 52.7 in July from 50.3 in June, rising to the highest level in 6 months. Good news: Russian manufacturing sector has now posted above-50 nominal readings in 12 consecutive months. Less bright news: Russian Manufacturing PMIs have signalled weak rate of recovery in 5 months to July and July reading was not quite as impressive as for the period of November 2016 - January 2017. Nonetheless, if confirmed in August-September, slight acceleration in Manufacturing sector can provide upward support for the economy in 3Q 2017, support that will be critical as to whether the economy will meet Government expectations for ~2% full year economic expansion.
  • Chinese manufacturing PMI gained slightly in July (51.1) compared to weak May (49.6) and June (504.), but growth remains weak. Last time Chinese Manufacturing posted PMI statistically above 50.0 (zero growth) marker was January 2013. This flies in the face of official growth figures coming from China.
  • India’s Manufacturing PMI fell off the cliff in July (47.9) compered to already weak growth recorded in June (50.9). Over the last 3 months, India’s Manufacturing sector has gone from weak growth, to statistically zero growth to an outright contraction.

Overall, GDP-weighted BRIC Manufacturing PMI stood at extremely weak 50.4 in July 2017, down from equally weak 50.6 in 2Q 2017. In both periods, BRIC Manufacturing sector grossly underperformed Global Manufacturing PMI dynamics (52.7 in July and 52.6 in 2Q 2017). Russia is the only country in the BRIC group with Manufacturing PMI matching Global Manufacturing PMI performance in July. Russian Manufacturing PMI was below Global Manufacturing PMI in 2Q 2017.

Net outrun: BRIC Manufacturing sector currently acts as a drag on global manufacturing growth, with both India and Brazil providing momentum to the downside for the BRIC Manufacturing PMIs.

2/8/17: Role of Clusters in empowering Human Capital: GCTV

My comment to GCTV on the importance of clusters in creating support base for human capital:

Saturday, July 29, 2017

28/7/17: 1H Marker: Russia on Track to a Weak Recovery in 2017

A quick top level update on the Russian economy from Bofit and Fitch Ratings.

Fitch Ratings today: “The recovery in Russia continues to gain traction. Domestic demand is responding to greater confidence in the economic policy framework, particularly as the inflation-targeting regime becomes entrenched. Activity in Turkey has bounced back rapidly from the coup attempt, with growth hitting 5% yoy in 1Q17. Momentum was supported by government incentives, including temporary fiscal measures and a jump in the Treasury commitment to the fund that backs lending to SMEs.”

Chart from BOFIT confirms the above:

Overall, the recovery is still on track, and remains gradual at best, posing elevated risks of reversals. For example, industrial output, having previously posted gains in January-May 2017, contracted in June 2017 on a quarterly basis. Still, industrial output was up 2% in 1H 2017 y/y.  Despite the U.S. and European sanctions, and generally adverse trends in the commodities sectors, mineral extraction sector expanded 3% y/y in 1H 2017 according to BOFIT. Oil output was up 2% and gas output was up 13%. The above figures imply that higher value added manufacturing posted sub-1% y/y growth in 1H 2017.

Agricultural production was basically flat - due, in part, to poor weather conditions, rising only 0.2% y/y in 1H 2017 and unlikely to post significant growth for FY2017 as crops reports are coming in relatively weak. That said, 2015-2016 saw record crops and very strong growth in agricultural output, so barring a major decline this year, agricultural sector activity will remain robust. Food production sector was the fourth highest growth sector over 2013- 1H 2017 period across the entire Russian economy, rising cumulative 17%  in 1H 2017 compared to 1H 2013 in real (inflation-adjusted) terms.

Construction sector posted a robust 4-5 percent expansion in 1H 2017 compered to 1H 2016, a rather positive sign of improving investment.

Pharmaceuticals (+36%), plastics (+25%), Chemical industry (+22%), paper industry (+19%) were the main sectors of positive growth over 2013-2017 period, according to data compiled by Rossstat.

Really good news is that household demand is now recovering. Retail sales by volume were up ca 1% y/y on a seasonally-adjusted basis and real disposable household income rose from the cycle lows to the levels last seen in May-June 2016. Bad news is that with income growth slower than retail sales and even slower than actual household consumption (which grew faster than domestic retail sales due to accelerating purchases abroad), Russian households are dipping into savings and credit to fund consumption increases.

We shall wait until July 2017 PMI figures come out over the next few days to see more current trends in the Russian economy, but overall all signs point to a moderate 1H and 3Q (ongoing) expansion in the economy, consistent with 1.2-1.3% real growth. The Economy Ministry recently reiterated its view that Russian GDP will expand at more than 2% rate in 2017. Achieving this will clearly require a large and accelerated cut in the Central Bank rate from current 9% to below 8%. Even with this, it is hard to see how above-2% growth can be achieved.

Agricultural and food production are quite significant variable in the growth equation. In 2016, Russia became number one exporter of wheat in the world, with annual production tipping 120 million tons - historical record. Bad weather conditions in 2017 mean that current expected output is estimated at around 17% below 2016 levels. Russia consumes 70 percent of its wheat output internally, so cuts to exports are likely to be on the magnitude of 1/2 or more in 2017. Domestically, food prices inflation is rising this year, threatening overall Central Bank target and putting pressure on CBR to stay out of cutting the key policy rate. Inflation rose in June to 4.4% - moderate by historical standards, but above 4% CBR target.

28/7/17: Risk, Uncertainty and Markets

I have warned about the asymmetric relationship between markets volatility and leverage inherent in lower volatility targeting strategies, such as risk-parity, CTAs, etc for some years now, including in 2015 posting for GoldCore (here: And recently, JPMorgan research came out with a more dire warning:

This is apt and timely, especially because volatility (implied - VIX, realized - actual bi-directional or semi-var based) and uncertainty (implied metrics and tail events frequencies) have been traveling in the opposite direction  for some time.

Which means (1) increasing (trend) uncertainty is coinciding with decreasing implied risks perceptions in the markets.

Meanwhile, markets indices are co-trending with uncertainty:
Which means (2) increasing markets valuations are underpricing uncertainty, while focusing on decreasing risk perceptions.

In other words, both barrels of the proverbial gun are now loaded, when it comes to anyone exposed to leverage.

Friday, July 28, 2017

28/7/17: Long term U.S. growth trend is still weak: 2Q 2017 Update

U.S. GDP growth estimate for 2Q 2017 came in at 2.6%, matching the post-1948 trend for expansionary periods almost to the notch. The problem, however, is that the trend is ... declining over time.

Here's the kickers to today's cheerful media reports on U.S. growth:

  1. Current expansion period average growth remains the shallowest amongst all post-recession recoveries since the end of WW2. That's right: the miracle of this Great Recovery is how weak it has been, despite all the Fed efforts.
  2. Current 4 quarters average for growth is 2.4%, which is only 0.2 percentage points above the overall recovery period average. Or, put differently, even before the revisions to 2Q 2017 numbers, last four quarters of growth have been un-inspiring. 
  3. The trend for historical growth during expansion periods has been sloping down since around the end of the 1980s. And we are, currently, still on that trend. In other words, recoveries are continuing to trend more anaemic over time.
So keep telling yourself that everything is coming out 'on expectations'. Just don't think about the pesky fact that expectations are trending lower.

27/7/17: U.S. labor markets are not in rude health, yet

As we keep hearing about the wonders of the U.S. labor markets, there is an uneasy feeling that the analysts extolling the virtues of the Great Non-recovery are bending the facts. Yes, unemployment is down significantly, and, finally, in recent months the participation rate started to climb up, although it remains depressed by historical norms. But these are not the only metrics of jobs creation or employment. Much overlooked are other figures, that paint a much less pleasant picture.

So with this in mind, lets update some of my old charts relating to the side of the labor markets than majority of analysts have forgotten to mention.

First up: average duration of unemployment. In other words, a measure of how long it takes for a person to get back into the job.

Good news is: the decline in duration of unemployment continues.  Better news: we are well past the crisis-period peak. Bad news: duration is at around 2009 levels, so not even at the levels pre-crisis. Worse news: current duration is higher than that recorded at the peak of any other recession in modern history. That's right: with miraculous recovery, we have folks collecting longer unemployment benefits than at the peak of any previous recession.

That was in absolute terms. Now, let's look at how we are performing relative to each pre-recession expansion:
Again, good news: the horror show of the peak during the height of the Great Recession is gone now. But, again, bad news: we are still at the levels of relative duration comparable to 15 months into the Great Recession. And, again, the worst news: after 108 months of 'recovery' we are much worse off in terms of duration performance than in any other post-recessionary recovery since 1948.

But what about employment, you might ask? Aren't U.S. companies generating huge numbers of jobs that are being filled by the American workers? Err... ok...

No. Employment is not performing well. Current cycle (from the start of the Great Recession through today) is long. But it is also extremely shallow when it comes to employment. So shallow, that it marks the worst long cycle in history (per above chart) and, when compared to shorter cycles, ... again, the worst cycle in history. 1953 cycle was bad - sharper jobs destruction than current, but it ended faster and on a higher employment index level than the current one.

So no, things are not fine in the U.S. labor markets. Not by the measures which are harder to game than standard unemployment stats.

Thursday, July 27, 2017

27/7/17: Designing a More Equitable System of School Access

In the decades old battle for the future minds, U.S. Republicans and Democrats have been constantly at odds when it comes to how one achieves, simultaneously, higher quality of education and more equitable access to education for those from less well-off families. In the mean time, one country - Chile - has been building up a system of vouchers that, after 36 years worth of experimentations, is delivering on both.

In 2008, Chile introduced a massive reform of its 27-years-old system of education vouchers by passing the Preferential School Subsidy Law (SEP). The system of education funding in Chile is based on universal school voucher payments, but until 2008, the system did not target explicitly those on lower incomes. Then, SEP changed this set up by hiking the value of the voucher by 50 percent for a large category of so-called “Priority students”, a category that primarily covers students “whose family incomes fell within the bottom 40 percent of the national distribution”.

A recent NBER study looked at the results (see full citation and link below).

Specifically, SEP reform stipulated that “to be eligible to accept the higher-valued vouchers from these students, schools were required to waive fees for Priority students and to participate in an accountability system.”

The study used data on math scores attained by “1,631,841 Chilean 4th-grade students who attended one of 8,588 schools during the year 2005 through 2012” and asked the following two questions:

  • “Did student test scores increase and income-based score gaps become smaller during the five years after the passage of SEP?” and
  • “Did SEP contribute to increases in student test scores and, if so, through what mechanisms?”

The study found that:

  1. “On average, student test scores increased markedly and income-based gaps in those scores declined by one-third in the five years after the passage of SEP.” So the effects were in desired direction for all students (improving outcomes) and stronger effect for targeted students (the Priority Students). Better result for all, more equitable result for those in most need.
  2. “The combination of increased support of schools and accountability was the critical mechanism through which the implementation of SEP increased student scores, especially in schools serving high concentrations of low-income students.” So poor performance of less well-off schools improved more than the performance of better-off schools. Again, better result for all, more equitable result for those in most need.
  3. Another important aspect of the study was to identify whether the new vouchers triggered a massive redistribution of better-performing poorer students away from less well-off school. In other words, whether the scheme reform benefited predominantly more those students from the less privileged background who would have gained otherwise. The authors found that “migration of low-income students from public schools to private voucher schools played a small role.”

So you can design a market-based solution for education system funding that does preserve schools choice, enhances educational outcomes for all, and reduces educational inequality.

Full citation: Murnane, Richard J. and Waldman, Marcus and Willett, John B. and Bos, Maria Soledad and Vegas, Emiliana, The Consequences of Educational Voucher Reform in Chile (June 2017). NBER Working Paper No. w23550. Available at SSRN:

27/7/17: The Gen-Lost is still lost...

Today, Marketwatch reported on a research note from Spencer Hill of Goldman Sachs Research claiming that the young workers cohorts in the U.S. have now caught up in terms of employment with older workers' cohorts.

Sadly, the argument is based on highly flawed analysis. The core data presented in support of this thesis is the unemployment rate, as shown in the chart below:

But official unemployment figures mask massive decline in younger cohorts' labor force participation rates, as evidence in this chart from Peterson Institute for International Economics:

In simple terms, when you reduce your employment base by moving people into 'out of workforce' category, you lower unemployment rate.  This is supported by other research, e.g. as reported here: Skewed, against the Millennials, workplace conditions are also to be blamed: or as highlighted in these data:


So, no, beyond superficially deflated official unemployment metric, there is no evidence of the labor force conditions recovery for the younger workers. The Generation Lost is still lost. And that is before we consider the life cycle effects of the crisis.

Wednesday, July 26, 2017

27/7/17: Work or Play: Snowflakes or Millennials?

Snowflakes or Millennials? Flaky or serious? Careless or full of determination? Attitudes or aptitudes? Well, here’s an interesting study on the younger generation.

“Younger men, ages 21 to 30, exhibited a larger decline in work hours over the last fifteen years than older men or women.” In other words, average hours of labour supplied have fallen for the younger males more than for the older cohorts of workers. Which can be a matter of labour demand (external to workers’ choice) or supply (internal to workers’ choice).

One recent NBER study (see below) claims that “since 2004, time-use data show that younger men distinctly shifted their leisure to video gaming and other recreational computer activities.”

So we have two facts running simultaneously. What about a connection between the two?

“We propose a framework to answer whether improved leisure technology played a role in reducing younger men's labor supply. The starting point is a leisure demand system that parallels that often estimated for consumption expenditures. We show that total leisure demand is especially sensitive to innovations in leisure luxuries, that is, activities that display a disproportionate response to changes in total leisure time.” Economics mumbo jumbo aside, the authors “estimate that gaming/recreational computer use is distinctly a leisure luxury for younger men. Moreover, we calculate that innovations to gaming/recreational computing since 2004 explain on the order of half the increase in leisure for younger men, and predict a decline in market hours of 1.5 to 3.0 percent, which is 38 and 79 percent of the differential decline relative to older men.”

Some data from the study:

So it looks like this data suggests that attitude beats aptitude, and choices we make about our recreational activities do cramp our decisions how much time to devote to paid work.

Full citation: Aguiar, Mark and Bils, Mark and Charles, Kerwin Kofi and Hurst, Erik, Leisure Luxuries and the Labor Supply of Young Men (June 2017). NBER Working Paper No. w23552. Available at SSRN:

26/7/17: Credit booms, busts and the real costs of debt bubbles

A new BIS Working Paper (No 645) titled “Accounting for debt service: the painful legacy of credit booms” by Mathias Drehmann, Mikael Juselius and Anton Korinek (June 2017 provides a very detailed analysis of the impact of new borrowing by households on future debt service costs and, via the latter, on the economy at large, including the probability of future debt crises.

According to the top level findings: “When taking on new debt, borrowers increase their spending power in the present but commit to a pre-specified future path of debt service, consisting of interest payments and amortizations. In the presence of long-term debt, keeping track of debt service explains why credit-related expansions are systematically followed by downturns several years later.” In other words, quite naturally, taking on debt today triggers repayments that peak with some time in the future. The growth, peaking and subsequent decline in debt service costs (repayments) triggers a real economic response (reducing future savings, consumption, investment, etc). In other words, with a lag of a few years, current debt take up leads to real economic consequences.

The authors proceed to describe the “lead-lag relationship between new borrowing and debt service” to establish “empirically that it provides a systematic transmission channel whereby credit expansions lead to future output losses and higher probability of financial crisis.”

How bad are the real effects of debt?

From theoretical point of view, “when new borrowing is auto-correlated [or put simply, when today’s new debt uptake is correlated positively with future debt levels] and debt is long term - features that are present in the real world - we demonstrate two systematic lead-lag relationships”:

  • “debt service peaks at a well-specified interval after the peak in new borrowing. The lag increases both in the maturity of debt and the degree of auto-correlation of new borrowing. The reason is that debt service is a function of the stock of debt outstanding, which continues to grow even after the peak in new borrowing.” It is worth noting a well-known fact that in some forms of debt, minimum required repayment levels of debt servicing (contractual provisions in, say, credit cards debt) is associated with automatically increasing debt levels into the future.

  • “net cash flows from lenders to borrowers reach their maximum before the peak in new borrowing and turn negative before the end of the credit boom, since the positive cash flow from new borrowing is increasingly offset by the negative cash flows from rising debt service.”

Using a panel of 17 countries from 1980 to 2015, the paper “empirically confirm the dynamic patterns identified in the accounting framework… We show that new borrowing is strongly auto-correlated over an interval of six years. It is also positively correlated with future debt service over the following ten years. In the data, peaks in debt service occur on average four years after peaks in new borrowing.” In other words, credit booms have negative legacy some 16 years past the peak of new debt uptake, so if we go back to the origins of the Global Financial Crisis, European household debts new uptake peaked at around 2008, while for the U.S. that marker was around 2007. The credit bust, therefore, should run sometime into 2022-2023. In Japan’s case, peak household new debt uptake was back in around 1988-1989, with adverse effects of that credit boom now into their 27 years duration.

When it comes to assessing the implications of credit booms for the real economy, the authors establish three key findings:

1) “…new household borrowing has a clear positive impact, and its counterpart, debt service, a significantly negative impact on output growth, both
of which last for several years. Together with the lead-lag relationship between new borrowing and debt service this implies that credit booms have a significantly positive output effect in the short run, which reverses and turns into a significantly negative output effect in the medium run, at a horizon of five to seven years.”

2) “…we demonstrate that most of the negative medium-run output effects of new borrowing in the data are driven by predictable future debt service effects.” The authors note that these results are in line with well-established literature on negative impact of credit / debt overhangs, including “the negative medium-run effect of new borrowing on growth is documented e.g. by Mian and Sufi (2014), Mian et al. (2013, 2017) and Lombardi et al. (2016). Claessens et al. (2012), Jorda et al. (2013), and Krishnamurthy and Muir (2016) document a link between credit booms and deeper recessions.” In other words, contrary to popular view that ‘debt doesn’t matter’, debt does matter and has severe and long term costs.

3) “…we also show that debt service is the main channel through which new borrowing affects the probability of financial crises. Consistent with a recent literature that has documented that debt growth is an early warning indicator for financial crises, we find that new borrowing increases the likelihood of financial crises in the medium run. Debt service, on the other hand, negatively affects the likelihood of crises in the short turn.”

In fact, increases in probability of the future crisis are “nearly fully” accounted for by “the negative effects of the future debt service generated by an increase in new borrowing”.

The findings are “robust to the inclusion of range of control variables as well as changes in sample and specification. Our baseline regressions control for interest rates and wealth effects. The results do not change when we control for additional macro factors, including credit spreads, productivity, net worth, lending standards, banking sector provisions and GDP forecasts, nor when we consider sub-samples of the data, e.g. a sample leaving out the Great Recession, or allow for time fixed effects. And despite at most 35 years of data, the relationships even hold at the country level.”

So we can cut the usual arguments that “this time” or “in this place” things will be different. Credit booms are costly, painful and long term.