1/1/2015: US Mint Gold Coins Sales: 2014

End of 2014 and Q4 2014, so time to update my relatively infrequent coverage of data for US Mint sales of gold coins. Here's the data for the sales of American Eagles and Buffalo coins.

Starting with quarterly data:

• Sales of US Mint gold coins in Q4 2014 reached 183,500 oz up on 141,000 oz in Q3 2014 and the highest reading since Q1 2014. However, y/y Q4 2014 sales were down 4.2% having posted a rise of 24.2% y/y in Q3 2014. There is quite a bit of volatility in Q4 sales. For example, Q4 2013 sales were down 29.5% y/y and Q4 2012 sales were up 74% y/y.
• Sales of US Mint gold coins also fell in terms of average coin weight. In Q4 2014, average coin sold carried 0.57 oz of gold per coin, down from 0.61 oz in Q3 and down from 0.71 oz/coin average in Q4 2013. Still, Q4 2014 reading was second highest in oz/coin sales terms in 2014.

Chart below illustrates.

Monthly trends were less favourable in December. Volume of gold sold via coinage sales by the US Mint fell well below the period average and the series have now been trending below historical averages (both across 2006-2014 range and 2012-2014 averages) since May 2013.

The same dynamics: falling oz/coin average, and falling number of coins sold can be traced in full year sales figures, as illustrated in the chart below.

As above clearly shows, the decline in total number of coins sold has been relatively moderate, compared to historical trend, with sales of 1,322,000 coins in 2014 running very close to 2006-2013 average of 1,361,625 coins. But sales in oz terms have been poor: in 2014 total sales of US Mint gold coins run at 702,000 oz against the 2006-2013 average of 983,250 oz. Thus 2014 was the third worst year on record (since 2006) in terms of sales of coinage gold, but ono the fifth worst year on record in terms of sales of coins by numbers. The average coin weight at 0.53 oz per coin in 2014 was the poorest on record.

Year on year full-year dynamics were poor as well: total coinage gold sold by oz fell 36% y/y in 2014 and there was a decline of 22% in the number of coins sold. Meanwhile price of gold declined (based on month-end USD denominated prices) by 9.94% y/y.

Most of the poor performance in US Mint sales took place in H1 2014, when coinage gold sales in oz terms fell from 790,500 oz in H1 2013 to 377,500 oz in H1 2014.

In the end, 2014 was a poor year for US Mint sales. Even stripping out the sales of the American Buffalo and looking at the American Eagle sales alone - thus allowing the data to cover 1986-2014 period - the trend remains to the downside for both oz sold and coin numbers, with oz sold under-performing the downward trend.

That said, sales of American Eagles remain above the averages for both coin numbers and gold volumes once we strip out 1998-1999 anomalies.

All in, the explanation for 2014 performance is continued decline in demand for gold coins from shorter-term investors seeking safe haven. In general, this is expected and is likely to continue: gold coins are normally the domain of collectors and longer-term long-only investors. We are witnessing a moderation in demand trends toward 1987-1997 and 2000-2008 averages. 

1/1/2015: Shared Liability: Debtor and Lender

In a recent blogpost on geography of Euro area debt flows prior to the crisis, I noted the extent to which Irish (and other peripheral euro area economies') debt bubble pre-2008 has been inflated from abroad (see here: http://trueeconomics.blogspot.ie/2014/12/27122014-geography-of-euro-area-debt.html). The argument, of course, is that the funding source, just as the funding user, should co-share in the liability created by the bubble.

This argument, advanced by myself and many others over the years of the crisis, has commonly been refuted by the counter-point that no such liability is implied: borrowers willingly borrowed from the banks, banks willingly borrowed from the markets (aka other banks) and that is where liability ends.

Here is a cogent paper on the subject from the Bank for International Settlements (not some lefty-leaning think tank or a libertarian hothouse of dissent): Turner, Philip, Caveat Creditor (July 2013). BIS Working Paper No. 419: http://ssrn.com/abstract=2384445).

The paper asserts that "One area where international monetary cooperation has failed is in the role of surplus or creditor countries in limiting or in correcting external imbalances." In common parlance, that is the area of liability of one economic system that, having generated surpluses of savings, provides funding to another economy.

"The stock dimensions of such imbalances - net external positions, leverage in national balance sheets, currency/maturity mismatches, the structure of ownership of assets and liabilities and over-reliance on debt - can threaten financial stability in creditor as in debtor countries." In other words, net lender (e.g. Germany) co-creates the imbalance with the net borrower (e.g. Ireland).

And thus, "creditor countries ...have a responsibility both for avoiding "overlending" and for devising cooperative solutions to excessive or prolonged imbalances."

Unless responsibility does not imply liability (in which case me being responsible for driving safely should not translate into me being liable for any damages done to other parties from failing to do so), we have confirmation of my logic: net lending countries (I refer you to the chart in the blogpost linked above) bear shared liability with the borrowers. By extension, lending banks share liability with the borrowers. Per BIS. Not just per the unreasonable myself.

1/1/2015: Population Ageing and Economic Growth

What happens to economic activity with population ageing? And, crucially, what happens in the context of free mobility of labour, currency union and open trade and capital mobility? These are the questions to be answered for European policymakers, facing rapid increases in population age and in some countries (Germany and Italy already) facing decreases in working age population.

An interesting paper on the subject was just published in the U.S. authored by Maestas, Nicole and Mullen, Kathleen and Powell, David, study titled "The Effect of Population Aging on Economic Growth" (October 2014, RAND Working Paper Series WR-1063: http://ssrn.com/abstract=2533260).

Per authors, "Population aging is widely expected to have detrimental effects on aggregate economic growth. However, we have little empirical evidence about the actual existence or magnitude of such effects. In this paper, we exploit differential aging patterns at the state level in the United States between 1980 and 2010. Many states have already experienced high growth rates of the 60 population, comparable to the predicted national growth rate over the next several decades. Furthermore, these differential growth rates occur partially for reasons unrelated to economic growth, providing a natural approach to isolate the impact of aging on growth."

The study predicts "the magnitude of population aging at the state-level given the state’s age structure in an initial period and exploit this predictable differential growth to estimate the impact of population aging on Gross Domestic Product (GDP) growth, and its constituent parts, labor force and productivity growth."

The result is an estimate showing "that a 10% increase in the fraction of the population ages 60 decreases GDP per capita by 5.7%. We find that this reduction in economic growth caused by population aging is primarily due to a decrease in growth in the supply of labor. To a lesser extent, it is also due to a reduction in productivity growth. We present evidence of downward adjustment of earnings growth to reflect the reduction in productivity."

31/12/2014: Irish Ghost Estates: The March of Zombies

An interesting recent article from the Irish Independent on the sad state of Irish ghost estates: http://www.independent.ie/business/personal-finance/property-mortgages/work-stopped-on-226-ghost-estates-across-ireland-30856439.html?utm_content=bufferd7e17&utm_medium=social&utm_source=twitter.com&utm_campaign=buffer

Interesting, from my perspective, not just in the fact that 226 ghost estates saw no work activity in 2014, despite the uplift in property prices and Government prioritising completion of ghost estates. But interesting due to numbers it revealed.

Take a deep breath: seven years after the crisis set in, and nine years after building activity contraction set in, Ireland (a country of 4.8 million inhabitants) still has 992 estates (as in multiple dwellings developments) unfinished. And of these, 776 estates have people residing on the site of abandoned construction. But that is not all, 271 more (on top of 992 above) are not completed, but deemed to have been 'substantially completed' (which can mean pretty much anything).

Good news, 1,854 ghost estates have been completed. Bad news is that the Year Four of Our Government's Recover Turnaround, only 271 ghost estates have been completed, which means that at current rate we are looking at 2017 or later before we are rid of the ghost estates. That is a decade of physical scars reminding us about less than a decade of excesses. Of course, given growth in homelessness, the rising spectre of banks repossessions, the social housing lists explosion and other fine mess, non-physical scars will be with us much longer.

31/12/2014: Falling Again: Russian PMIs for December

HSBC and Markit released Russian PMIs for December, showing deteriorating conditions in Russian economy, as expected, given the severe Ruble crisis that hit mid-December.

Manufacturing activity posted a reading of 48.9 which is down from 51.7 in November, signalling a switch from a rather average growth to a contraction. December reading was close to being statistically significant for a sharp decline. Q4 2014 average Manufacturing PMI was at 50.3 which is better than Q4 2013 reading of 50.0 and worse than Q4 2012 reading of 51.7. But December figure breaks three consecutive months of above 50.0 readings and Q4 2014 reading is now below Q3 2014 average of 50.8.

Services PMI continued sub-50 print for the third consecutive month, coming in at 45.8 in December. Q4 2014 showed sharp deterioration in Services compared to Q3 2014 (50.2), as well as compared to Q4 2013 (53.0) and Q4 2012 (56.8).

Composite PMI fell to 47.2 in December from already weak 47.6 in November, marking third consecutive month of sub-50 readings. Q4 2014 average is at 48.0, far worse than Q3 2014 average of 51.1 and well below Q4 2013 average (51.1) and Q4 2012 average (52.7).

Overall, as chart above clearly shows, the downward trend in Russian economic activity across all sectors, the trend that set in around November 2012 and started flashing signals of recessionary dynamics around Q4 2013, remains in place.

31/12/2014: Ruble Crisis: Banking System in a Shut-Down Mode

Something for Russia analysts to watch comes January 12: The CBR will be offering RUB 1.1 trillion in 3-mo repo auction with eligible collateral being lowered to allow non-marketable assets. That is roughly USD20 billion in one go.

Meanwhile, Russian CB has been bailing out banks in line with the announcement made two weeks ago and passed via an emergency legislation by the Duma. Trust Bank was the first one to get a bailout of RUB99 billion in a form of 10-year loan and additional RUB28 billion loan for "Otkrytie" - financial intermediary that will take over Trust Bank. But the bailout is a bit of a misnomer here. Instead, it is a backdoor QE. "Otkrytie" already announced that it will spend RUB99 billion it borrowed from the CBR at 0.51% pa, to buy Russian Federal bonds. On back of that, S&P downgraded "Otkrytie" confirming rating of BB-/B, but moving it to negative outlook.

This was followed by the recapitalisation for VTB. The Government approved RUB250 billion funding for VTB which will be paid into two tranches. The first one of RUB100 billion was already deposited with the bank and the second one is forthcoming in Q1 2015. With both tranches in place, VTB CT1 capital ratio will be expected to rise to 12% from current 10.2%. VTB got the first tranche on the following terms: 30 years deposit at inflation+1% margin per annum, calculated every 6 months and payable every 6 months.

In reality, here's what's happening on the ground. 2014 has been marked by freezing of external funding sources (due to sanctions), rising corporate demand for credit (due to sanctions) and delletion of deposits. Deposits inflows were predominantly forex, demand for credit was predominantly in Rubles. The crisis is made worse (worse probably than 2008-2009 one) because capital buffers of the banks are weaker, relative to regulatory benchmarks and funding sources were more reliant on external funding and were shorter term. The CBR drive to reduce number of banks competing for dwindling deposits base has been not aggressive enough, so market fragmentation is still a problem: too many banks with

The banking crisis is now being compounded by the breakdown in payments systems. In September 2014, the CBR facilitated setting up of the new National Platform for Payments Cards (NPSK) that is supposed to become operational by march 1, 2015. Interestingly, this week the CBR published a list of 50 major or significant payments providers operating in Russia - a list that excludes both Mastercard and Visa.

The recaps will continue on. National Wealth Fund is set to inject ca RUB394 billion (10% of the fund value) into the systemically important banks, namely banks with own capital in excess of RUB100 billion, the list of which includes only Sberbank, VTB, Gazprombank, Rosselkhozbank, Alfa-bank, VTB-24, Bank of Moscow, Unicredit Bank and Rosbank. The injection is supposed to be used for infrastructure investments by the banks. Funds will be disbursed in the form of deposits and in debt paper issued to fund infrastructure investments.Cost of funding will be set at the rate at which the NWF will provide deposits to the banks. Banks will report quarterly on funds use.

Basically, we are witnessing a system that is heading into a major crisis - the hatches are being welded shut, not just battened. Whether that takes place before the flaring up of the next bout of Ruble crisis or not will determine how 2015-2016 are going to play out.

30/12/2014: Who Owns Government Debt?

An interesting chart via DB, mapping sovereign debt holdings across the advanced economies:

As the chart clearly shows, Irish Government debt is disproportionately held in the Central Bank. Other countries with similar proportion of CB-held debt - UK, US and Japan - all deployed direct QE. Ireland, of course, deployed virtually the same QE-like stimulus predominantly to the IBRC.

Another interesting feature is the share of Government debt held by foreign agencies: roughly 20% of the total, or 11th lowest in the sample of 21 countries. That is pretty low, given the amount of PR-talk the Government has been deploying around foreign buyers of Irish bonds.

In contrast, predictably, we rank the third after Greece and Portugal in the share of Government debt held by foreign official sector. This will decline once the IMF 'repayment' is finalised. Domestic banks' holdings of Irish debt are third lowest in the sample, and domestic non-banks holdings are 5th lowest. This is unlikely to change, given the sheer quantum of Government debt outstanding, relative to the overall economy's capacity and demand, and given the low yields on Government debt being generated.

The kicker of all of this is that owing to years of mismanaged bailouts, we are now saddled with the legacy of rescuing private debt holders in the banks. This legacy is simple: instead of private debt we have official debt, held predominantly by official sectors and our own CB, guaranteed by the Irish State. In other words, more of our debt is now super-senior in both rights and default terms.

29/12/2014: Historical Evidence on the Size of Fiscal Adjustments

A recent IMF paper looked at the historical precedents of large scale fiscal adjustments across advanced and emerging economies in the aftermath of the major fiscal crises. Escolano, Julio and Mulas-Granados, Carlos and Terrier, G. and Jaramillo, Laura paper titled "How Much is a Lot? Historical Evidence on the Size of Fiscal Adjustments" (IMF Working Paper No. 14/179. http://ssrn.com/abstract=2519005) argue that "the sizeable fiscal consolidation required to stabilize the debt-to-GDP ratios in several countries in the aftermath of the global crisis raises a crucial question on its feasibility."

To answer this question, the authors look at historical evidence "from a sample of 91 adjustment episodes of countries during 1945-2012 that needed and wanted to adjust in order to stabilize debt to GDP."

"We find that in most cases fiscal adjustment is sizeable and the debt-to-GDP ratio stabilizes by the end of the episode, albeit at higher levels. In at least half of the episodes, countries managed to improve their primary balance by 5.4 percent of GDP (4.8 percent of GDP in cyclically adjusted terms). The sample distributions of the levels and changes in the primary balance (actual and cyclically adjusted) show that, while there are significant differences across advanced and developing countries in terms of the levels of primary balances achieved, the changes in primary balances are comparable across the two groups."

"The fiscal adjustment implemented was enough to close the primary gap in two-thirds of the episodes. This implies that debt stabilized, and in most cases was put on a downward trend." Given the hope-inspiring dynamics above, however, the follow-up is less impressive: "This does not however imply that debt returned to initial levels. While countries kept primary balances well above those observed before the adjustment episode, they did not sustain primary balances at the highest levels for prolonged periods of time. This suggests that countries make substantial efforts to stabilize debt but, once this is achieved, they see room to ease primary balances and do not necessarily seek to get back to the lower initial debt-toGDP ratio."

 "We find that consolidations tended to be larger when the initial deficit was high and adjustment efforts were sustained over time." In addition, "Several factors are found to be significantly associated with the size of fiscal adjustments. …The results also show that fiscal adjustment tended to be higher when accompanied by an easing of monetary conditions (as measured through a reduction in short-term interest rates) and, to a lesser extent, an improvement of credit conditions (measured as the change in credit to the private sector as a percent of GDP), especially in advanced economies."

Couple of figures. In the below,
CAPB: cyclically adjusted primary balance as a percent of potential GDP;
CAB: cyclically adjusted balance as a percent of potential GDP

Note the following interesting facts:

• Ireland's fiscal adjustment post-2009 has been shallower than its adjustment post-1986. 
• The cause of this shallower adjustment was the collapse of the credit markets in Ireland plus the on-going deleveraging of the real economy, not present in 1986 crisis. Also, the factors not accounted for in the list presented in the chart. In 1986 episode such factors were positively contributing to fiscal adjustment. In 2009 episode - they had negative impact. We can only speculate what these factors might have been, but clearly they are not related to external trade, or FDI. Which suggests they were domestic.
• Ireland's adjustment was longer in the 1986 episode than in 2009 episode, but that is because the paper does not go beyond 2012. And the adjustment post-2009 episode is not completed still, even in 2014.
• CAPB is the main driver of adjustment in 2009 episode, and is much larger than in 1986 episode. 
• Ireland's fiscal adjustment since 2009 has been shallower than that of Greece since 2008, Portugal since 2010 and Spain since 2009, although it has been longer running that in Portugal and as long running as in Spain. In fact, the UK - a country that lent funds to Ireland for adjustment - is running similar magnitude fiscal adjustment as Ireland since 2009. A bit rich for us to be claiming to have taken most of fiscal pain in this crisis.

So what does the above tell us about Euro area peripherals' adjustments? IMF paper says that things tend to go well when:

1. adjustment efforts were sustained over time, which suggests we are in for a much longer run than the Government's 'free from IMF' meme suggests;
2. there is an an accompanying easing of monetary conditions, which we do have, courtesy of the ECB, except it is unclear how does this relate to the cases similar to the current crisis where monetary accommodation is simply fuelling asset bubbles and temporarily relieving mortgages pain, while doing nothing for growth; and
3. to a lesser extent, by an improvement in credit conditions, which is yet to materialise, 5 years since 2009.

Not that any of the above will pause the IMF public statements about sustainability of adjustments everywhere and anywhere.

29/12/2014: Some Tough Data from Moscow: Growth & Ruble

Couple of reminders that the Russian crisis is not over, yet.

November GDP figures show GDP down 0.5% y/y - the first month of decline since October 2009. In October 2014, growth was +0.5% and 0.1% m/m. November m/m posted a decline of 0.2%. All figures hereinafter are seasonally adjusted and working day adjusted.

Decline in November was driven by a broad range of sectors: industry, construction, private services, taxes on goods and import duties. Extraction sectors, electricity, water & gas, and retail sales posted positive growth.

Investment fell 1.9% m/m in November. This is before the December currency crisis and two massive interest rates hikes. So expect more red ink here when December figures come out.

January-November overall GDP growth is now down to 0.6%. Seasonally-adjusted construction sector activity was also revised - Q1 2014 posted a revised decline of 2.5% y/y against previous estimate of 2.3%, Q2 2014 posted a decline of 1.6% from an estimated decline of 0.7%, Q3 2014 decline was 0.4% from previously estimated rise of +0.1%. In October, construction sector posted revised m/m growth of 3.5% and in November the sector activity fell 1.5% m/m.

Industrial production posted another month of poor results. In September, industrial activity grew at 1.4% m/m, following by slower 0.2% growth in October, and a decline of -0.9% m/m in November.

Real personal disposable incomes fell 4.6% in Q1 2014, rose 3.3% in Q2 2014 and 1.7% in Q3 2014. In November, real disposable income fell 2.9% m/m, having posted growth of 2.4% m/m in October. We can certainly expect further and deeper declines in December on foot of massive increases in interest rates and a drop in Ruble valuations.

Retail sales rose 0.2% in November, after posting 0% growth in October.

Unemployment remains unchanged at 5.2% - the rate that has been steady for the last 6 months.

Exports of goods in November reached USD37.6 billion representing a decline of 19.7% y/y and 7.1% drop m/m. Imports of goods in November stood at USD23.2 billion, down 22.1% y/y and a decline of 13.5% m/m. So despite sharper decline in imports in percentage terms, trade balance deteriorated from USD17.0 billion in November 2013 to USD14.4 billion in November 2014.

Full year trade figures estimates are charted below:

As suggested on this blog, imports declines are expected to run deep: 8.4% y/y in 2014 against expected exports decline of 3.0%. The result is the forecast increase in trade surplus of USD11.5 billion or 7.0% y/y.

Largely unrelated to the above news, Ruble seems to have reversed some of the gains made last week and is down some 9.4% against USD and Euro on lower oil prices:

Credit: @Schuldensuehner:

Euro chart:

The point is - last week's measures are temporary in their nature:

• 17% interest rates cannot be sustained without completely demolishing banks balancesheets, companies and households;
• Convincing larger corporates to cut their forex deposits will have at most short-run effect, as demand for forex is likely to ramp up due to debt maturity in Q1 2015.
• Threats of investigations and pursuit of 'speculative transactions' will do preciously nothing as most of trade will be coming via ex-Russia intermediaries.
• Massive spike in interventions from the CBR are not sustainable, given the pressures on reserves from the banks, debt redemptions and corporates/investment.

Meanwhile, core weaknesses in the economy and oil price dynamics will remain. And with them, there will be continuous pressure on the Ruble.

28/12/2014: Oil Prices: USD70bbl in 2015?.. May be so...

Forecasting oil prices is a rather tough job, especially if we are witnessing a regime change, rather than a temporary glitch in the markets. Nonetheless, here is the set of forecasts from the markets (options), forecasters (consensus) and Goldman Sachs:

Noticeable features: all, but futures markets bets are starting to converge in H2 2015. End-of-2015 forecasts are pretty close for Goldman and Bloomberg survey.

A bit more detailed reporting on forecasts here: http://uk.businessinsider.com/r-oil-prices-likely-to-rebound-in-second-half-of-2015-reuters-poll-2014-12?r=US.

All in, it does appear that USD70 bbl average price for 2015 is not out of line with market consensus, albeit it is not consistent with futures markets contracts.

28/12/2014: Health Impact of Self-Employment

A recent paper by Zhang, Ting and Carr, Dawn, titled "Does Working for Oneself, Not Others, Improve Older Adults' Health? An Investigation on Health Impact of Self-Employment" (American Journal of Entrepreneurship, Volume 7(1), pp. 142-180, 2014. http://ssrn.com/abstract=2512600) "examines the health impact of being self-employed versus working for others among older adults (aged 50 ) and its implications".

Economic reasoning behind the study is straightforward:

1. As authors note, "facing an aging workforce, self-employment at older ages may provide an economic benefit via an alternative to retirement.
2. As authors do not note, self-employment is generally associated with higher levels of stress, induced by income uncertainty, volatility, lack of proper scheduling of vacations and breaks, tax-induced anxiety and other factors that can potentially adversely impact self-employed.
3. Self-employment in older age is becoming increasingly the likeliest prospect for future employment for many workers, especially as economies gear toward services sectors with fast depreciation of skills and increased specialisation.

Despite all of the above, as stated in the paper, "little research has examined the health effects of self-employment in later life."  Zhang and Carr study "comprehensively examines health using a 29-item index to measure the impact of self-employment status on changes in older adults' overall health." The authors provide control for "potential endogeneity and simultaneity issues."

The study finds that "self-employment compared to wage-and-salary jobs result in better health, controlling for job stress and work intensity, cognitive performance, prior health conditions, socioeconomic and demographic factors. This positive self-employment impact stands out in knowledge-based industry sectors. In labor intensive industry sectors such as Durable Goods Manufacturing, self-employed older adults' more gradual retirement seems to result in a health advantage over wage-and-salary employees."

These results are quite interesting from both microeconomic and macroeconomic perspective. Self-employment as opposed to full retirement secures more significant pensions cover in older age, coincident with poorer health and greater demand for healthcare. It is also, it seems, reduces cost of healthcare in the first years of retirement. In addition, self-employment of older workers suggests that ageing demographics impact on aggregate growth and productivity growth can be mitigated in the Western societies, if there is an improved system of incentives for older workers to transition into retirement more gradually, over longer time.

28/12/2014: Unhappy Cities

According to a new paper published by NBER, "there are persistent differences in self-reported subjective well-being across U.S. metropolitan areas, and residents of declining cities appear less happy than other Americans. Newer residents of these cities appear to be as unhappy as longer term residents, and yet some people continue to move to these areas." The question is why?

"While the historical data on happiness are limited, the available facts suggest that cities that are now declining were also unhappy in their more prosperous past. One interpretation of these facts is that individuals do not aim to maximize self-reported well-being, or happiness, as measured in surveys, and they willingly endure less happiness in exchange for higher incomes or lower housing costs. In this view, subjective well-being is better viewed as one of many arguments of the utility function, rather than the utility function itself, and individuals make trade-offs among competing objectives, including but not limited to happiness."

While this sounds very plausible, an interesting follow up question is: what happens to new movers when they get better offers elsewhere or once they retire or their employment terminates? Do these newcomers leave? Do they attempt to secure new employment in the area? Do they engage in entrepreneurship whilst in their employment or after?

The reason why these questions are pivotal is that human capital is like other forms of capital: once it is mobile, it moves to higher returns on investment, but it also is footloose. Since investment in human capital does not end with current stage employment, loss of past human human via exit from the area is also a loss of future human capital increases. Securing human capital is about as important as attracting it.

For the paper quoted, see: Glaeser, Edward L. and Gottlieb, Joshua D. and Ziv, Oren, Unhappy Cities (July 2014). NBER Working Paper No. w20291. http://ssrn.com/abstract=2471184