Economics 05/06/2010: Economics of Fiscal Stimulus

This is an unedited version of my article for June-July issue of the Village Magazine.

Weeks into a new round of ‘talks’ over the public sector reforms and Ireland’s Policy Kindergarten squad is getting more agitated by the issue of cuts in the Government expenditure. The logic of their arguments, led by the likes of Tasc, the Irish Times, and an army of Unions-employed ‘economists’, is perverse: “In order to get the economy back on track, we need to borrow more and spend on public services and wages.”

There are three basic arguments why stimulating Irish economy though increased public spending won’t work in the current conditions even in theory, let alone in practice. These are: the structural nature of the fiscal crisis we face, the size of the debt we face, and the lack of evidence that stimulus can work in a country like Ireland.

Structural deficits

Economists distinguish two types of deficits: cyclical and structural. The first type of deficits occurs when a temporary economic slowdown leads to an unforeseen decline in revenue and acceleration of certain components of spending (e.g. unemployment insurance and social welfare). By its definition, the cyclical deficit will be automatically corrected once economy returns to its long term growth path.

In contrast, structural deficits are those that arise independently of the short term changes in economic growth. They are the outcome of unsustainable increases in permanent spending and/or decline in the long term growth potential that might arise from a severe crisis.

In the case of Ireland, both of the latter factors are at play. Various estimates of the extent of structural deficits carried out by the likes of IMF, OECD, the European Commission, ESRI and independent analysts range between one half and two thirds of the 2009 General Government deficit, or 7-9.5% of GDP.

Reckless expansion of Government spending in the period of 2001-2007 is the greatest cause of these – not the collapse of our tax revenue. In the mean time, our economy’s long-term growth rate has declined from the debt-and-housing-fueled 4.5% per annum to a Belgium-like 1.8% per annum.

In 2000, General Government Structural Balance stood at roughly -0.5% of GDP. By 2008 this has fallen to almost -11% courtesy of a massive build up in permanent staff increases in the public sector, rises in welfare rates, explosion in health spending and creation of a gargantuan army of quangoes and supervisory organizations.

Forget, for a second, that majority of these expenditures represented pure waste, delivering nothing more than top jobs for friends of the ruling class, plus scores of jobs for public and quasi-public sector workers. Between 1981 and today Ireland has recorded not a single year in which Government structural balance was positive. Windfall stamps, VAT and capital gains tax receipts over 2001-2007 have masked this reality, as Goldman Sachs structured derivatives masked the reality of Greek deficits.

We are not getting any better

Over the recent months, the Government has been eager to ‘talk up’ our major selling points. Ireland, it goes, is a country with stabilized public finances and low debt to GDP ratio.

Last month, Eurostat exposed the lie behind the ‘stabilized public finances’ story. It turns out our Government has decided to sweep under the carpet billions of cash it borrowed in 2009 to recapitalize Anglo. Courtesy of this, our deficit for 2009 was revised to a whooping 14.3% of GDP – topping that of Greece.

But Irish General Government deficit this year is expected to come in between 11.7% and over 12% of GDP, depending on who is doing the forecasting – Department of Finance or ESRI. And this is before we factor in March 2010 statement by the Minister for Finance, promising over €10 billion for the banks this year. This means that, as the rest of the world is coming out of the recession, our fiscal deficit for 2010 is expected to either match or exceed the revised level achieved in 2009. Some stabilization.

Irish Government debt is expected to reach 78-82% of GDP by the end of 2010 – on par with Eurozone’s second sickest economy, Portugal. With Nama and banks recapitalizations factored in, Irish taxpayers will be in a debt hole equal to between 117% and 122% of GDP by 2011 and to 137% by 2014. At the point of the Greek debt crisis implosion last year, Greece had second highest debt to GDP ratio in the EU at 117%, after Italy with a massive 119%.

In totality, current crisis management approach by the Irish State is going to cost every Irish taxpayer in excess of €117,000 in added tax liability. Neither Iceland nor Greece come close.

Economy on steroids

Still think that we should be stimulating this economy through more borrowing?

Take a look at the private sector debts. In terms of external debt liabilities, Ireland is in the league of its own amongst the advanced economies. Our overall debts currently are in excess of the critically high liabilities of the HIPCs to which we are sending intergovernmental aid. And rising: in Q3 2009, our external debt liabilities stood at a whooping USD 2.4 trillion, up 10.8% on Q3 2007. Of these, roughly 45% accrue to the domestic economy – more than 6 times our annual national income.

Ireland’s share of the world debt is greater than that of Japan and more than double that of all BRICs combined, once IFSC companies are included. Over the next 5 years, the entire Irish economy will be paying out around €206,000 per each taxpayer in interest on this debt. Adding more debt to this pile is simply unimaginable at any stage, let alone when the cost of borrowing is high and rising.

These figures show that the main cause of the current crisis is not the lack of liquidity in the system, but an old-fashioned problem of insolvency.

This problem is directly related to the actions of the Irish state. Over the last decade, there was a nearly 90% correlation between the average increases in the Irish tax revenues plus the rate of economic growth and the expenditure growth on capital and current spending sides. In effect, courtesy of the ‘Boom is getting boomier’ Ahearn/Cowen team Ireland had two bubbles inflating next to each other – a private sector borrowing bubble and a public sector spending one. Government’s exuberant optimism, cheered on by the Social Partners – the direct beneficiaries of this ‘fiscal policy on steroids’ approach – explains why during Brian Cowen’s tenure in the Department of Finance, Irish structural deficit doubled on his predecessor’s already hefty increases.

But what went on behind the glossy Exchequer reports was the old-fashioned pyramid scheme. Some got rich. Temporarily, we had an army of politically connected developers and bankers stalking the halls of premier cars dealerships and property auction rooms.

Permanently, an entire class of public employees reaped massive dividends in terms of shares in privatized enterprises that cumulated in their pension plans. Current claims that because the values of some of these payoffs have declined over time (often due to the intransigent nature of the unions in the semi-state companies, staunchly resisting change and productivity enhancing reforms) is irrelevant here. Prior to their privatization, these companies were called 'public' assets. Creation of any, no matter small or large, private gains to their employees out of the companies' privatizations or securititization through pensions funds liabilities of their assets in favor of employees, therefore, is nothing more than an arbitrary, unions-imposed grab of the public asset.

Benchmarking, lavish pensions and jobs security – also paid out of the economy leverage (just think of the NPRF - explicitly created to by-pass the illegal, under the EU rules, taxation of economy for provisioning for future public sector pensions liabilities) – was a cherry on top of the cake. Public companies management got dramatically increased pay and a permanent indemnity against competition through a regulatory system that was all but a client of their semi-state companies.

From our hospital consultants to our lawyers, academics and other professionals – a large army of state-protected, often non-competitive internationally professional elites collected state-subsidised pay so much in excess of their real productivity that we became the subject of diplomats’ jokes.

Our state’s response to this was telling. Just as the country was borrowing its way into insolvency, our Government gave billions to aid developing nations. That was the price our leaders chose to pay to feel themselves adequate standing next to Angela Merkel and Nicolas Sarkozy at the EU summits. Incidentally, as the country today is borrowing heavily to cover its basic bills, Brian Cowen still sends hundreds of millions of our cash to aid foreign states and has recently decided to commit over €1,000 million – full year worth of the money he clawed out of the ordinary families through income levies – to the Greek bailout package.

Economics on Steroids

Still think more state-centred economy is the solution to our problem? Irish economists, primarily those affiliated with the Unions are keen on talking about the ‘positive multiplier’ effect of deficit-financed stimulus. Sadly for them, there is no conclusive evidence that borrowing at 5 percent amidst double-digit deficits and ‘investing’ in public services does any good for the economy.

Firstly, one has to disregard any evidence on fiscal stimulus efficiency coming out of the larger states, like the US, where imports component of public and private expenditure is much smaller than in Ireland. The US estimates of the fiscal stimulus multiplier also reflect a substantially lower cost of borrowing. Even if Ireland were to replicate US-estimated fiscal stimulus effects, higher cost of our borrowing will mean that the net stimulus to Irish economy will be zero on average.

Second, international evidence shows that for a small open economy, like Ireland, the total fiscal multiplier effect starts with a negative -0.05% effect on economic growth at the moment of stimulus and in the long run (over 6 years) reaches a negative -0.07-0.31%. Add the cost of financing to this and the long-term effect of deficit financed stimulus for Ireland will be around -2.3% annually.

Third, no one on the Left has a faintest idea what the new spending should be used for. Simply giving borrowed cash to pay the wage bill in the public sector would be unacceptable by any ethical standards. Any investment that is bound to make sense would have to focus on our business centre – Dublin, where infrastructure deficit is acute and potential demand is present. Alas, this will not resolve the problem of collapsed regional economies. Pumping more cash into the ‘knowledge economy’ absent actual knowledge infrastructure of entrepreneurship, private finance, skills and without a proven track record of exporting potential, is adventurist even at the times of plenty.

In short, the idea that expanded deficit financing will support any sort of real recovery in the economy is equivalent to arguing that pumping steroids into a heart attack patient can help him run a marathon.


Ireland needs severe rethinking and reforms of the grossly inefficient and ethically non-sustainable spending and management practices of our public sectors. It should start with significant rationalization of expenditure first and then progress to a more deeply rooted revision of the public sector objectives and ethos.

Ireland also needs a significant deleveraging of what is a basically insolvent economic structure. This too requires, amongst other things, a significant reduction in overall public spending. Far from ‘borrow to spend’ policies advocated by the Left, we need ‘cut to save’ policies that can, with time, yield a permanent increase in the national savings rate, productive private investment and improved returns on education and skills. Otherwise, we might as well give our college graduates a one-way ticket out of Ireland with their degrees, courtesy of Tasc and the Unions.

Economics 04/06/2010: Bond markets are still jittery

For all the EU efforts:

• Throwing hundreds of billions into the markets in bonds supports;
• Banning 'speculative' transactions;
• Talking tough on reforms;
• Bashing rating agencies into a quasi-submission; and
  
• Proposing a 'markets calming' [more like 'markets killing'] financial transactions taxes

There has been preciously little change in the way the bond markets are pricing sovereign debt of the PIIGS. More ominously, the crisis is not only far from containment, it is spreading. Following PIIGS, the attention is now shifting onto BAN countries - Belgium, Austria and Netherlands. And in the case of Austria, the unhappy return of the Eastern European woes is now seemingly on the cards.

How so? Look no further than Hungary. The country had taken IMF bailout money, promising to deliver severe austerity measures. It now faces a new round of pressures due to once again accelerating deficits. It looks like the cuts enacted were not structural in nature, amounting to chopping capital expenditure programmes rather than current spending... Sounds familiar? so here we go again (courtesy of Calculated Risk blog): spreads are rising (Ireland's position as the second sickest country by this metric remains unchallenged) and CDS rates are rising as well (Ireland's still in number 3 spot).
As Calculated Risk points: "After declining early last week, sovereign debt spreads have begun widening for peripheral euro area countries. As of June 1, the 10-year bond spread stands at 503 basis points (bps) for Greece, 219 bps for Ireland, 195 bps for Portugal, and 162 bps for Spain."

Let's get back to Hungary, though: yesterday, Hungarian officials said that instead of 3.8% of GDP deficit target, 2010 is likely to see the deficit widening to 7-7.5% of GDP. Who's to blame? Well, per Reuters report: '"fiscal skeletons" left by the previous Socialist administration'.

Economics 02/06/2010: Regional variations in labour markets

While working on a project relating to economic policies in Ireland, I was compiling data on regional variations in various series. Here is a set of interesting graphs detailing the labour force differences across the main regions.

Each data set reflects the latest available QNHS data through Q3 2009 and each is presented in two charts - full history and a snapshot of the crisis dynamics since 2007.

First the unemployment rates
Notice that since time memorial Dublin runs at or below average in terms of unemployment rates. This pattern is no persistently broken, with Dublin unemployment performing remarkably better in this crisis. Also note that the top tier of unemployment black spots in the country also remains relatively resilient over time. This has to put to test any assertion that state policies to deal with longer term unemployment are working.

Take a look at a closer time frame, relating to the current crisis:
You hear a lot about the MNCs and exporting companies holding our line from a total collapse of economy. Well, say the same for Dublin, South-West and Mid-East. Of course, the latter is largely, hmmm, Greater Dublin, really.

The chart above also hints at something more disturbing here. Recall that early rounds of layoffs impacted predominantly construction sector and associated services. Well, look at Midlands and South-East. It does appear that the two regions were experiencing significantly faster rates of jobs losses in the early parts of the crisis than any other region.

One wonders what is the exact distribution of jobs in the country relative to places of residence. This, of course, is a long running question that CSO is refusing to ask on QNHS. What trouble can there be, folks, in asking a recipient to state where they physically work. It would tell us a lot about people's commuting patterns (helping to better plan transport systems) and about where people are actually employed (helping us to better plan associated business services provisions). But no - CSO staunchly refuses to ask. Why? Because the state is most likely unwilling to admit that the National Spatial Strategy and the IDA/EI mandates to produce jobs for regions is failing. Ireland has natural hubs of jobs and jobs creation potential - Greater Dublin area, Cork area, Galway-Limerick area. This is where jobs concentrate and where companies want to be. So how about a challenge to CSO - ask an important question, will you? Have some gumption...

Back to data: labour force participation rates next
What the charts above show is the precipitous decline in labour force participation rates since the peak of H2 2007. And these declines are worrisome, for we normally tend to ascribe the destructive effects of the economic crises to unemployment, forgetting about those who leave the work force altogether. Well - take a look at charts above.

Another disturbing realization on the foot of the above charts is that regions with lowest participation rates also tend to be regions with higher unemployment. This is important because it signals that even in a small country like Ireland, mobility between residential location and work location is still restricted (by distance, lack of proper roads, transport shortages etc). It also suggests that in the long run, areas with higher unemployment tend to become traps for non-participation in labour force. The vicious spiral of being jobless in an area with no jobs creation leads to becoming disillusioned and dropping out of the work force for good.

And this implies higher rates of overall dependency. Remember - these are numbers for able bodied adults. So if we take the rate of unemployed and add to it the rate of those who are not in the labour force, we get a proportion of population that needs someone else to work for them to sustain themselves. Now, a caveat here - of course some of those who are not in labour force are gainfully engaged in work at homes - non-market activities that are productive and include, among other very important ones, like carrying for the elderly or ill, raising children etc. These, however, are not the majority in these numbers. Nor are they likely to be distributed predominantly into higher dependency areas of the country. So conclusions presented below stand:
Predictably, the lowest dependency ratios are in high work regions: Mid-East and Dublin. And although these ratios rose in these two regions through the crisis, they are still well below the national average and leagues below the dependency ratios for the likes of Border and South-East regions. Here's a closer look:
Of course, what these trends mean is that throughout the entire series duration (from 1998) Dublin and Mid-East have acted as a subsidy generating regions for the rest of the country. Someone had to pay for the higher dependency rates in regions that are above country average (since the welfare rates are not varied geographically).

Economics 02/06/2010: Live Register - no longer flatlining

Liver Register numbers are out today, erasing much of the optimism that might have been building up about unemployment figures over the last couple of months. Here are the updated charts:
Unemployment rate is now rising again, reaching 13.7% after staying flat for the last four months at 13.4%. My forecast in January was that we will hit 13.5% in May. I was wrong. Well, not as wrong as some of our 'official' forecasters who were saying that we are turning the corner on unemployment...

Let's put today's news into perspective:
Notice 'missing' bubbles before May? That is because we had no growth in LR figures (they were actually shrinking) in February-April. Since the beginning of the year, mom, LR increased by 5,800 in January, fell by 2,300 in February, rose by 600 in March and contracted by 500 in April. With May increase of 6,600, we now have a net addition to the LR over the first 5 months of the year of 10,200. The Exchequer cost of these will be around €325 million per annum. In fact it will be probably higher as contractions in employment most likely have taken place in higher value added sectors, given that construction sector has virtually no jobs left to lose.

Taking a slower snapshot of the LR:
You can clearly see an uptick in the series in May. Given how relatively 'sticky' (trend-driven) LR is, this suggests that we might be heading for a new acceleration.

Next consider average weekly series. These are not seasonally adjusted:
However, taking seasonally adjusted data and extracting monthly series shows that we are firmly above the 'stabilization' line of zero change in the LR mom:
So here you have it - labour markets beg to differ from the Government's official line that 'all indicators point to a recovery'...

Economics 02/06/2010: Central bank data analysis

Latest monthly data from the CB is out and here are a couple of updates on series I've been covering before.

First harmonized competitiveness indicators (EU-wide data update coming soon):
Notice some serious progression on competitiveness front is finally starting to take place. This is good. The trend is also good - strong downward trajectory in the series since November 2009. Accelerating again since March. Data lags should not be this significant, so I will be keeping a watch on earnings data from the CSO.

For all the good news, so far we are still in the zone of low competitiveness, down to March 2006 level and well above the period when Ireland Inc was performing at much stronger rates in the 1990s. Remember, these are real indicators, so price levels changes since the 1990s are factored in already.

Private sector credit. First the totals:
We are back to August 2007 levels and the fall rate is slowing down. Year on year change, subsequently, is flat at -9.3% same as in March. Too early to call it a recovery or even a full stabilization, as seasonality suggests that we might see some trend reversal in the short run. Remember, these are declines on already bottom-hitting 2009!

Next: mortgages.
Levels are down to July 2008 and the rate of decline is -1.6% yoy, compared to -1.4% in March. This, however, can be due to a significant declines in mortgages due to write-offs of defaulting loans. In addition, this deterioration rate might be also masking the fact that pretty much anyone in distress who could have done so has already re-negotiated their mortgages in 2009. Thus, only the really tough cases are still sitting out there.

The data on actual new borrowing is below. At the aggregate levels, there is no turn around in household investment, which, of course, is the main leading indicator of recoveries. Also worrisome is the fact that there is no deleveraging of mortgages debt.

Private sector credit outside mortgages is dynamically virtually identical to the total private sector credit figure reported above. Year-on-year changes seem to be reflective of some seasonal effects, with improved rate of contraction in April. General trend is for flatter rates of decline overall since about January. This means little, however, as we need a term structure decomposition of credit in order to tell if this is really a flattening of the downward trajectory or simply restructuring of non-performing lines of credit.

Now, let's take a look at actual changes in rates and volumes in PS credit. First, new loans:
Notice that both for corporates and households, longer term rates are moving up, while shorter term rates are moving down. This likely reflects banks' and interbank credit markets' expectation for a steepening in the interest rate curve, plus some easing in wholesale cost of credit in March. Also note that mortgage rates for new, and especially for fixed rates, mortgages are rising. Hardly a robust support for the housing market.

On corporate investment side, sizable declines for short term maturity loans - operating capital, and reasonably improving environment for larger investment-suitable loans with longer term fixes.

On volumes side, there is a worrisome increase in all shorter term loans - a sign that both companies and households are reliant increasingly on short bank finance for operational and short-term credit. This might mean two things:

1. These increases might reflect increase in supply against a pinned up demand; or
2. These increases might be consistent with increased cash flow pressure on companies (if non-payment and defaults by clients is rising) and households (if arrears are building up on the side of unemployed and underemployed after the households have gone through their savings and redundancies).

We can't tell which one of these forces is operative here. But it does not look to me like operational demand is rising naturally. Remember, so far we only have strong exports performance across the economy. This means you would expect an increase in trade credits (short-term). Most of trade finance in Ireland is actually not done via Irish banks, but through MNCs-own global arrangements. Apart from exports, it is hard to see where organic demand for short term loans would come from.

An even more interesting picture is emerging when we look at existing clients:
Notice how all of the rates changes (except for 5 year plus maturity corporate loans) are trending up? Are the banks ripping off their existent client base to beef up their margins? Well, lets put these changes side by side:
Notice that the above table comparisons are really only loose approximations. But there is a remarkable regularity with which existent loans holders are being loaded with the almost opposite type of changes in rates charged as compared with new clients.

Economics 1/6/2010: Numbers game at Anglo

Last night, I sat down to run through possible scenarios for the Anglo's 'The Bad and the Ugly' Banks division. You see, something was telling me right off the start that the idea of a 'Good' Bank just doesn't really square off with our knowledge of the bank's operations to date.

So I posited to myself the following question: given Nama transfers and rumored split off of €12-15bn worth of loans into a 'good' bank, can the resulting entity be viable? Like a scientist in a lab, I donned on a white coat (well, really my favorite UofChicago sweatshirt), pulled out a Petri dish (my Excel) and started observing the split of that outright not-so-beautiful and very toxic (to the taxpayers) bacteria, called Anglo...

Here are the results, first in numbers and then in plain English:

Step 1: recall we have pumped €10.3 billion worth of promisory notes into the bank alone. Relying on my yesterday's analysis (see details here), I reproduced the demand that a 'Good' Anglo will generate for funding these promisory notes. Now, a reminder - these numbers (penultimate column) correspond to interest only charge on Anglo from the promisory notes. They exclude principal repayment and other recapitalization funding already in the bank.
Bah, I said, the thing in the Petri dish of mine looks pretty ugly. Ugly as in unable to cover the taxpayers-due interest on capital it receives at the first glance.

Ok, I said to myself, but may be were the new 'Good' bank to grow over time, it will become relatively viable with time? Suppose the 'Good' bank generates no impairments going forward (unrealistic assumption, but suppose it does), suppose that 'Good' Anglo grows its book at 5% (generating no new impairments). Further suppose that there's some value in the 'Bad' bank - so assume 20% of the loans transferred to it perform in the future (an extremely optimistic assumption, but what the h***ll, not much out of line with the general assumptions the Government has been making all through its management of the crisis).

The question I asked then was: with all these rosy assumptions in place, what amount of interest payments annually can Anglo afford?

To compute this, I took several scenarios:

1. I allowed 'Good' Anglo to take €12 or €15 billion in loans on board;
2. I assumed that it generates 2% of the loan book annually (another optimistic assumption - as it corresponds to an efficiently operating bank in terms of costs, book of business and funding costs - all of which are not exactly characteristic of the Anglo)
3. I then assumed three different potential burden levels on interest (recall, no principal) repayment at 30% of the total annual return by the bank, 25% and 20%. Let me explain here that a 30% number is utterly unrealistic, implying that almost a third of the entire operating revenue of the bank will be used to pay interest on a small share of its capital funding. This will, in effect, leave no surplus to pay bonuses (of any kind) and dividends (of any kind) as well as to finance bank's insurance etc. 25% mark is also unrealistic, while 20% is back-breaking for a bank, but can be probably sustained over a couple of years.

Table below shows the results by stating the amount of interest repayment that the bank can generate across both its 'Good' and 'Bad' divisions. Blue-bold numbers mark the first time that the annual interest funding requirement gets met.
All of this is fine, I said to myself next, but before the interest requirement is first met on the annual basis, there are years of the bank not covering the interest bills. These will cumulate.

My next question, therefore was: How soon can the bank break into the 'black' vis-a-vis interest repayment alone?
Table above shows the cumulated interest arrears from the €10.3 billion in promisory notes. It clearly shows that under all scenarios, save one (the most optimistic scenario) the entire Anglo operation cannot be expected to generate enough cash to cover even the portion of its interest bill. In fact, under the more realistic scenario (last two columns), Anglo - 'Bad' and 'Good' combined - will continue to accumulate interest arrears on the taxpayers funds (ex €4 billion in direct capital it received) through 2020.

There is no principal repayment charge in the above, nor is there a chance of receiving anything close to the interest bill, even assuming that we do not roll up interest on the cash we put in. In simple words - the entire Anglo operation is so fundamentally bust, that the taxpayer is likely to never receive even a few cents on the euro of the money we've put into it.

The only thing that grew in my Petri dish was a voracious bacteria capable of hoovering taxpayers money at a speed unimaginable to any other bank.

One wonders if that is what Mr Alan Dukes and our Government mean when they are saying that proceeding with keeping Anglo on a respirator amounts to minimizing the cost to the taxpayers.