Friday, March 22, 2013

Repaying Mortgages in BOI

The annual reports of Bank of Ireland gives some insight into the changes in mortgage balances on their accounts. The 2012 interim report has tables that report the balance outstanding on mortgage accounts (both owner-occupier and buy-to-let) by year of origination at the end of 2011 and the end of 2012.  See section on mortgages beginning on page 321.  The figures in Table 2 from both years are reproduced here.

BOI Mortgage Balances

For example, using the first row of figures we can see that at the end of 2011, BOI had €123 million of outstanding mortgages which had originated before 1996.  The Dec 2012 column shows that repayments during 2012 reduced the outstanding balance on these mortgages to €94 million.  Thus €29 million or 23.6% was paid off the balances of these mortgages.

In 2012, the total amount of mortgage credit issued in Ireland by BOI fell by €369 million or 1.3%.  However, if we account for the €981 million of mortgages that originated from 2012 this means that of €27,854 million of mortgage debt outstanding at December 2011, €1,350 million was repaid during 2012.

The reduction in the outstanding balance could be due to:

  • repayments on the existing loan
  • re-mortgages to a new loan or new provider
  • write-downs on the existing loan

It is likely that the first of these is the main source of the balance reduction in BOI mortgages.  This means that around 4.9% of the outstanding mortgage balance that BOI had at the end of 2011 was paid off during 2012.

On a simple straight-line basis this would mean that the stock of debt could be fully paid off in 20 years, however, over time the capital portion of repayments will increase as the interest due falls (due to capital repayments). 

This would suggest that the average duration of BOI’s mortgage book given the current repayment pattern is around 15 years.  This is pretty standard.

Of course, averages can be misleading.  There will be some mortgages which will be repaid much quicker than that, many others which will require much longer and others which will likely never be repaid at all.

The report doesn’t provide the amount of arrears owing on the mortgage accounts.  What we can see is that during 2012 around €1.3 billion of capital repayments were made on BOI mortgages.  During the same period mortgages with a total balance of €0.9 billion fell in arrears of more than 90 days, bringing the total to €3.6 billion.  The amount of missed payments (capital plus interest) is significantly smaller than the amount of capital payments made.

Monday, March 18, 2013

Ten Questions on the Cypriot Deposit Tax

Although it was still came as a surprise when announced the possibility of a depositor contribution to the Cypriot rescue has been in the offing for some time.  The €16 billion rescue programme for Cyprus is in some ways similar to the Irish programme.  Ireland is in an €85 billion programme (55% of GDP), but €17.5 billion came from our own resources.  Cyprus is considering entering a €16 billion programme (90% of GDP), but must find €6 billion from its own resources.

Ireland used the National Pension Reserve Fund (NPRF) for our contribution.  Cyprus has no such sovereign wealth fund and has to look elsewhere to find the €6 billion.  The €6 billion could have been found through a long-term, low-interest loan from the the ESM which could be serviced, and possibly repaid, through increased taxes in the future.

Instead the proposed approach is for a “once-off” levy on all bank deposits in Cyprus to raise the €6 billion now.  The proposed tax is set at:

  • 6.75% of all bank deposits up to €100,000, and
  • 9.9% of all bank deposits in excess of €100,000

But these are still subject to change.  The proposal raises a number of questions.  Here is a set of ten.

  1. Why is it called a tax?
  2. Why does it apply equally to all banks?
  3. Why does it apply to all customer deposits?
  4. Why does it apply to deposits below €100,000
  5. Why are all deposits above zero subject to the cut?
  6. What about inter-bank deposits?
  7. Will it be applied per account or per depositor?
  8. What about the bondholders?
  9. What about the Russians?
  10. Is it a good measure?

Some brief discussion of these is below the fold.

1. Why is it called a tax?

Technically it is called “an upfront one-off stability levy” on deposits but it is not a tax or “fiscal measure” as claimed by Olli Rehn.  The revenue does not accrue to the Cypriot government.  It is a type of debt-for-equity swap.  The banks will swap the appropriated value from each depositor for an equivalent value of shares in the banks.  So, someone with €10,000 on deposit on Friday will have €9,325 in their account on Tuesday (or Wednesday or Thursday or whenever) and will be given €675 of worth of shares in the bank.  Existing shareholders will see their much-devalued holdings diluted and it is very unlikely that the new shares will maintain their starting nominal value, but there is some talk of linking the swap to future revenues from natural resources (or something).

2. Why does it apply equally to all banks?

Just like Ireland, Cyprus has a set of banks some of whom have made big losses and others who have made more moderate losses.  The Irish banks lost money by getting deposits from abroad and lending them out to people (concentrated in the real estate and construction sectors) who could not repay it.  The Cypriot banks lost money by accepting deposits and using some of the them to buy around €15 billion of Greek government bonds.  When the EU mandated that Greece had to default on its sovereign bonds some Cypriot banks endured significant losses.  The tax applies to all deposit instruments held in Cyprus so includes banks which incurred these losses and those that didn’t.

3. Why does it apply to all customer deposits?

The tax applies to demand accounts, notice accounts and current accounts.  Timing means that someone paid monthly who received their salary into the bank account last Thursday or Friday will lose around 1/15th of this month’s pay under the current proposals, whereas someone paid early next week will get their full salary.

4. Why does it apply to deposits below €100,000?

Cyprus has a €100,000 Deposit Protection Scheme.  The DPS “is activated in the event a decision is reached that a member bank is unable to repay its deposits”.  However, none of the Cypriot banks are saying they are unable to repay their deposits (yet).   This is not a failure by the banks to repay deposits, it is mandated reduction in deposits by the government. [Can all deposit guarantee schemes be circumvented in this way? It would seem so.]  Of all the customer deposits in Cypriot banks it is estimated that €30 billion are covered by the DPS.

5. Why are all deposits above zero subject to the tax?

A lower cut-off of even €20,000 would likely remove huge numbers of people from being subject to this tax without significantly reducing the amount of money to be raised by it.  A deposit of €20,000 implies a tax liability of €1,350.  The reduction from such a move would probably be less than 10% of the total expected to be raised by the levy.

6. What about interbank deposits?

Almost all the attention has been on the €68 billion of retail deposits held by banks in Cyprus, but these banks also had €24 billion of deposits from monetary financial institutions (MFIs).  These inter-bank deposits will not be subject to the deposit levy.  There is also liquidity from the ECB and ELA from the Central Bank of Cyprus.  The monthly balance sheet of the central bank suggests that around €9 billion of ELA has already been forwarded to Cypriot banks.

7. Will it be applied per depositor or per account?

A depositor with €200,000 in one account will be faced with a €16,650 reduction in their balance whereas someone with 4 x €50,000 deposits could see €13,500 lifted from their accounts if done on a per-account basis.  Will the banks/government know how much a person has on deposit across all the banks?

8. What about the bondholders?

For a start there are very few bonds issued by the Cypriot banks.  At the end of February the total was €1.7 billion (compared to €68 billion for customer deposits).  The Eurogroup statement includes a reference to a “a bail-in of junior bondholders” but it is not clear how much that will generate.  There is close to no senior unsecured bonds in the Cypriot banking system.  Any claims that depositors are being sliced to save bondholders is well wide of the mark.  There are no bondholders to save so such an argument is a red herring.

9. What about the Russians?

There are lots of claims that one of the reasons for the depositor bail-in is to target Russian money that has fled to Cyprus.  I think this is confusing stocks and flows.  There is some Russian money in Cyprus but the flow of money in and out is much much greater.  Money that was on deposit in Cypriot banks last Friday will be subject to the levy but the vast amount of money that has moved in and out over the past few years is unencumbered by this.  The move will probably significantly reduce the flow of Russian funds through Cyprus.

10. Is it a good measure?

In theory, any step in the direction of having the creditors of a bank carry the consequences for losses in excess of the capital of that bank is to be welcomed.  If depositors give money to a bank and that bank loses it (for whatever reason) then it should be the depositors who bear the losses.  In theory of course. 

The issue is muddied when we try to distinguish between “big” depositors who can afford to know better and “small” depositors who are simply not in a position to examine the finer details of a bank’s balance sheet (even if that was a reliable statement of the riskiness of a bank’s position).  Bondholders should evaluate such risks but they are absent in this case. 

Much as it may be to the chagrin of many people banks are different, and instances where providing public support to ailing banks are not difficult to comprehend.  This was not deemed possible in Cyprus because of the initial level of government debt (70% of GDP at end 2011) and the decision to focus on the bank’s creditors followed that.  [Aside: What would have happened if Ireland’s government debt was 70% of GDP in 2007 rather than 25%?]

This step will have a negative impact on the Eurozone’s banking system.  The scale and scope of this will only emerge over the coming weeks and months.  It may be limited to Cyprus, or it may not.  It is possible that this move will have a positive impact on the Eurozone’s sovereign bond markets.  In countries with (f)ailing banks there is now the precedent that creditors of the banks will carry the burden rather than sovereign bondholders.  This precedent may not be repeated and the contents of the Eurozone’s playbook are as difficult to predict as ever.

So, yes continue with this deposit haircut but introduce a lower limit of €20,000 per person to exclude working balances in personal current accounts and “small” depositors.  Cyprus will have to put together a plan to significantly downsize its banking sector.  The ECB will have to provide significant amounts of liquidity to cover the deposit flight.  Ireland and the ECB did so in 2010 and we have slowly pulled back from the abyss.  But wouldn’t it have been better if the Anglo depositors had carried the can?

Public Sector Pay Cuts

Here is a table of possible pay changes under for some stylised sample public servants under the proposed public sector pay agreement.

CP II Pay Changes

The figures were produced by Joseph Byrne and Sons Actuaries for the 24/7 Alliance.  Without the specifics for all of the sample cases it is hard to analyse the figures.  The calculation for one of the cases (the 3.6% reduction for a Garda Sergeant at the top of the scale hired before 1995) is below the fold.

More interest in Cyprus

Here is a quick snapshot of retail interest rates in Ireland, Cyprus and for the Euroarea average.

Ireland Cyprus Interest Rates

Consider a saver who had €10,000 in January 2010 and has placed it in accounts with an agreed maturity (of up to 2 years) since then.  Over the three years the Irish saver would have earned a 3.1% average rate of interest and, net of DIRT (which has increased from 25% to 33% over the period), would have around €10,700 in the account today.

A saver in Cyprus using similar accounts would have earned 4.2% on average on their deposit and, net of the 15% Cyprus Defence Levy applied to interest income, would have had €11,100 in their account last week.

If the proposed “stability levy” at 6.75% is applied to this balance it would reduce it to €10,400, some €300 less than their Irish counterpart.  The interest rate and tax differentials would see this difference eliminated before the end of next year.

When compared to the Euroarea average it is likely, given the interest rates above, that the outcome over the three years since 2010 for the Cypriot saver even with the proposed levy would be close to that for the Euroarea average.

And how about inflation over the three-years?  Here is the average of the annual HICP inflation rate since January 2010.

  • Ireland: 0.6%
  • Euroarea: 2.3%
  • Cyprus: 3.0%

Inflation was highest in Cyprus but is not substantially below the Euroare average.  The lower rate for Ireland is mainly driven by the price deflation experienced up to the end of 2010.

Of course, this mechanical approach suggesting Cypriot savers aren’t doing too bad is utterly irrelevant and completely ignores the wider reality of targeting deposits (and particularly guaranteed deposits) in this manner but that is getting plenty of coverage and providing much speculation elsewhere.

Sunday, March 17, 2013

“Ireland is not Cyprus”

We haven’t heard this beauty yet, but no doubt as the week progresses the classic “Country A is not Country B” will be rolled out.  The Cypriot developments of the past few days are very significant.  All bank deposits as of Friday 15th March will be subject to “an upfront one-off stability levy applicable to resident and non-resident depositors” per the Eurogroup statement.

The €5.8 billion to be raised from the levy is equivalent to 32% of Cypriot GDP (2012: €18 billion) or 8.5% of the deposits in the Cypriot domestic banking system.  For comparison, 32% of Irish GDP (2012: €162 billion) would be more than €50 billion and 8.5% of the deposits in Irish banks (including IFSC) would be €21 billion.

Deposits in the Cypriot domestic banking system are 382% of GDP; the equivalent figure for Ireland is 152%.  The Cypriot banks have a lot of deposits.  Here is a table of Irish and Cypriot bank deposits (excluding deposits from other banks in both cases).

Cyrpus Ireland Deposits

Around 62.5% of the deposits in the Cypriot banking system are labelled as “Resident” but the private sector numbers are “high” when compared to the Irish figures.  Both Cypriot households and non-financial corporations have greater than 2.5 times more deposits in their banks relative to the Irish equivalents in GDP terms.

The population of the Republic of Cyprus is around 875,000 so the €26.3 billion of deposits from the household sector imply a per-capita figure of around €30,000.  In contrast the 4.6 million population of Ireland implies a per-capita household sector deposit of €20,000.  GDP per capita in Cyprus is €20,000; GDP per capita in Ireland is €35,000.

Per-capita household sector deposits in Cyprus are 150% of per-capita GDP.  Per-capita household sector deposits in Ireland are 57% of per-capita GDP (or 71% of per-capita GNP).  The official statistics indicate that the Cypriots have a lot of money in the bank.

Some figures on the distribution of household deposits would be great.  Data through the income distribution for the median and inter-quartile range would do but it is impossible to determine that from central bank statistics.

Maybe the Cypriots do have a lot of money in the bank.  The DIRT enquiry highlighted the prevalence of bogus non-resident in accounts in Ireland through the 1980s and 1990s.  What odds on a prevalence of bogus resident accounts in Cyprus now?  Maybe Ireland and Cyprus are not so different after all.

Wednesday, March 13, 2013

Who benefits from Mortgage Interest Relief?

The Statistical Report of the Revenue Commissioners has a chapter on Income Distribution Statistics.  Within that Table IDS15 gives some details on the relief given for interest paid on home loans – Mortgage Interest Relief (MIR).  Here are the totals from that table.

Mortgage Interest Relief Distribution

The total amount of MIR allowed in 2010 was just under €280 million, with an average relief of €850 given to the 328,000 or so recipients.  The amount has fallen in recent years and was as high as €700 million in 2008 when the ECB rate peaked at 4.25%.  It has since fallen to 0.75%.

The final column shows that the amount of relief granted generally increases with income and the highest average going to those with an income over €275,000.  Although there are limits to the amount of relief that can be claimed, it is clearly a function of the size of the loan.   Those on higher incomes, on average, have bigger loans.

Around €20.5 million of relief is granted to Income Tax cases with an income over €100,000, though issues of jointly-assessed Income Tax returns, separately-assessed married couples and the tax case to which the relief deter definitive judgements.

It can be noted that at €279 million the amount of relief granted is less than the amount expected to the collected by the Local Property Tax this year (€250 million), and that even next year, when the full rate applies, the average MIR of €850 will be in excess of the property tax bill that these mortgaged home-owners will face.  MIR is due to be phased out by 2017.

Effective Tax Rates

The tax system and its nuances is a frequent topic in the public discourse.  A fertile area for such discussions are the headline, marginal and effective income tax rates.  In particular, emphasis is put on the Personal Income and Corporate Income tax rates. 

The headline rates of Personal Income Tax are 20% and 41% and the headline rate of Corporation Income Tax is 12.5%.  The top marginal tax rate, including USC (7%) and employee PRSI (4%) is 52% for individuals.  Non-PAYE earners face a marginal rate of 55% as there is an additional 3% USC levy on self-employed income over €100,000, while public sector workers above €60,000 have a marginal tax rate of 63.5%, inclusive of the ‘pension’ levy.

The Annual Statistical Reports from the Revenue Commissioners provide lots of useful information on effective tax rate.  These are taken from the Income Tax and Corporation Tax elements of the 2011 report which provide figures for the 2010 tax year.

Effective Tax Rates 2010

The figures for Personal Income refer just to Income Tax paid at the 20% and 41% rates and exclude PRSI and the Income and Health Levies which were also paid in 2010 as well as other Income Taxes such as DIRT. 

The pattern of effective Personal Income Tax rates with income are reasonably well understood.   The effective Income tax rate is zero up to around €17,000 and remains below 10% up to €40,000.  As income rises it rises to around 25% for incomes above €125,000 with the highest effective income tax rate around 30%. (PRSI and USC are additional to this.)  Details using preliminary 2011 data are in this table.

Here is a table that shows the pattern of the effective Corporate Income Tax rate through the range of Net Trading Income, which is found after certain allowances (losses carried forward, capital allowances and other deductions) have been subtracted from Gross Trading Profit.

Effective Corporation Tax Rate

Aside from companies with negative or very small net trading income (< €25,000), the lowest effective corporation tax is for companies reporting a net trading income of €10 million or more.   

There are around 45,000 companies in total in the figures and 32,000 are in the first two categories above.  There are around 500 in final category and 1% of companies paying corporation tax contribute 77% of the total amount paid.  These companies pay an average amount of €7.5 million of Corporation Tax.

Why does €70.0 billion of gross trading profits (and another €0.8 billion of balancing charges) yield only €4.2 billion of corporation tax?  There are various adjustments that are made to gross profit from any year to get the taxable profit for that year.  Here is a summary of the tables from the Revenue’s release.

Corporation Tax Deductions

Any proposals to raise the effective Corporation Tax rate without changing the headline rate of 12.5% need to address the deductions set out above.  Here is a similar table for Personal Income Tax, this time from the Revenue’s report on Income Tax.

Income Tax Deductions

Thursday, March 7, 2013

The Property Tax as a Progressive Tax

The passing of the legislation for the (Local?) Property Tax and the announcements from the Revenue Commissioners of how they intend to collect it have once again put this relatively minor issue towards the top of the public debate.

The now-abolished €100 Household Charge was expected to raise €160 million in 2012 from the 1.6 million liable properties.  In 2013, it is expected that the half-rate Property Tax will raise €250 million – an increase of €90 million.  This is an average annual increase of €56 per household or a little more than €1 a week.

Is the Property Tax progressive?  Without detailed household income and property valuations it is difficult to say but an broad overview of income by tenure status suggests it is.  This is a table from the 2009/10 Household Budget Survey.

Income and Tenure Status

[This is data up to 2010.  The recently released Survey on Income and Living Conditions put gross weekly household income at €1,015 in 2011, a 1.1% drop on the above figure.]

Owner-occupiers have higher incomes than other categories.  The group who own their home outright have an average annual gross household income (direct income plus state transfers) of nearly €47,000, while owner-occupiers with a mortgage had an average annual gross household income of €75,000.  If is, of course, the case that averages only provide a partial picture and a fuller insight into the distribution would be better.  Alas it is not available.

Households living in rental accommodation have lower gross incomes than owner-occupiers and are not legally liable for the Property Tax (though determining the economic incidence of the tax is difficult).

There are around 1.15 million owner-occupied homes in Ireland, roughly even divided between those who own their home outright and those who have a mortgage on their home.  There are around 450,000 households in rented accommodation with 150,000 renting from a local authority or voluntary body and the other 300,000 households renting from a private landlord.

Finally, here is an extract from a table in the SILC on disposable income (gross income after direct taxation) though it doesn’t break down home-owners into mortgaged/non-mortgaged households.

Disposable Income and Tenure

Printfriendly