Ronan Lyons | Personal Website
Ronan Lyons | Personal Website

November 2017

Should we ban house-building inside the M50?

The housing woes of Ireland – and Dublin in particular – at this stage well document and largely well understood. There are few who think we are fighting the last war. As the ESRI said last week, the housing market not undergoing the kind of credit-fuelled bubble that afflicted it in the early 2000s.

Nonetheless, just because it is not a credit bubble doesn’t mean that it’s a healthy housing market. Far from it. New CSO figures out this week confirm that the stock of dwellings, measured in euro terms, had been largely stagnant the past decade. While this may have made sense for the period 2007-2012 – when the country as a whole had too much, rather than too little housing – the lack of new housing supply over the last five years has cost the country.

This cost has a number of elements. Probably the most visible cost to society is rising homelessness – especially of lower-income families. Another long-term cost is the impact the lack of housing is having on competitiveness, especially in Dublin. The American Chamber of Commerce recently published a report, which I was involved in, highlighting the challenges faced by their members as they seek to expand and create new jobs in the city.

But I was involved in another report – for Activate Capital, a financer of developers joint-funded by the taxpayer through ISIF. This report found that the bulk of Ireland’s housing need is not actually for the type of housing we are used to seeing built.

There are significantly more family homes in Ireland than there are families: almost 25% too many, when you compare the 700,000 families in Ireland with the 900,000 dwellings to fit families. This is not even a problem in the greater Dublin area, where there is a 10% surplus.

Instead, the country’s housing shortage is entirely driven by apartments, or homes for 1-2 persons, and not by 3-4 bedroom houses. The term ‘apartment’ is a bit narrow. Ireland’s roughly 500,000 missing apartments does include medium-to-high rise apartments in urban cores for young professionals – and low-to-medium rise suburban apartments for empty-nesters.

But Ireland’s missing homes also includes tens of thousands of units in purpose-built student accommodation and co-living developments for key workers and young professionals. It also includes a similar number of missing homes in our largely non-existent independent living and assisted living segments for the country’s older residents.

These half a million missing homes represent a backlog – but it is not even a simple matter of meeting the backlog and then returning to ‘business as usual’ and building lots of family houses. When it comes to Ireland’s demographics, the outlook for the rest of the twenty-first century will be driven by three main forces: population growth, urbanisation and falling household size.

Unlike any other European country, Ireland will experience faster population growth in the 21st century than in either the 19th or 20th. For the century and a half to 1990, Ireland typically lost 5% of its population each decade. During the 21st century, the country will gain 5%. Every other European country is experiencing a slowdown.

You could argue that, since the Famine, the country has been a late bloomer, not only in terms of living standards but also population densities. This argument holds true for urbanization also. Ireland is now roughly as urban as the typical Western European country was 50 years ago. Over the next half-century, the country will go from 65% urban to 80% or more urban.

And lastly, there is household size. Once again, we are behind the curve. All European countries are on a journey from 4 persons or more in the typical household to just 2, or maybe slightly above. Ireland remains well above our peers, with 2.8 persons currently in the typical household. But this number has been falling steadily in Ireland since the 1960s. And it will continue to fall over coming decades.

Add these three factors together, and by 2080, you have a population of perhaps 6.3 million people living in Ireland, 80% or more of whom living in the cities and with the vast majority of households comprising just one or two persons. This is not a recipe for strong demand for housing estates even further out from our urban cores.

This is a recipe for building hundreds of apartments – of whatever type – in Dublin and throughout the country, every month for decades. Just taking Dublin alone, the city will in rough terms 2,500 apartment blocks with an average of 200 apartments built over the next half-century.

With a need for apartments this size, it certainly seems perverse of local authorities to favour commercial – especially office space – over residential. And it also seems perverse to allow development of estates of family houses when the city has an abundance of them already and when there is such a scarcity of apartments.

Given this stark background, it would seem prudent for Dublin’s four local authorities come together and agree both on a long-term multi-decade goal for housing in the city. It also seems wise that residential development, where it happens in the city, should be primarily – almost exclusively – in the form of apartments, not houses.

==

An edited version of this post was originally published in my column in the Sunday Independent.

The scale of the rental market challenge in Dublin

The Daft.ie Rental Report out earlier this week found that rents are still rising at double-digit rates around the country. Both nationally and in Dublin, this is the sixth quarter in a row that rents have been at least 10% higher than a year previously.

Outside Dublin, rents are now roughly one half higher than their lowest point, which was in early 2012. In Dublin, the increase has been even greater – rents have risen by three quarters from their lowest point in late 2010.

While South County Dublin, and Dublin 14 and 16, have seen rent increases of about two thirds, in Dublin 8, rents have increase by more than 90% from their lowest point. It is likely that within the next six months, one of the Dublin postcode – possible Dublin 1, 7 or 8 – will have seen rents double within eight years.

A natural reaction for many to this is to demand limits on rent increases. This is akin to a Minster for Health banning people from having high temperatures. High rents, like high temperatures, are not the problem – they are the symptom of the problem.

Fair enough, you might think, but I’ll still take a paracetamol when the need arises, thank you very much. And if that’s all that controls on rent increases did, then there would little to worry about.

The problem is that controls on rent increases turn the market into a system with insiders pitted against outsiders. Those who have a lease keep it. Those who have to move, either into the city or to somewhere new because circumstances change, lose out. To extend the paracetamol analogy, would you take one if it meant that your headache would go off and afflict someone poorer than you instead?

To solve the underlying problem, rather than just the symptom, the whole country – but Dublin in particular – needs significantly more supply of rental homes. The last time rents rose rapidly, in 2006 and early 2007, there were an average of 10,000 rental homes in Dublin posted on daft.ie each quarter.

Over the following five years, there were a variety of different totals posted each quarter and taking the picture as a whole, it reveals that, when roughly 14,000 Dublin homes were listed in a three-month period, rents were stable. More than 14,000 meant that rents fell (2008-2010 in particular), while less than that total meant rents rose.

In late 2012, the total number of listings in a three-month period fell below 14,000 and it has continued to fall since. Since the start of 2014, there has been an average of just 7,500 Dublin homes posted for rent every three months. It is therefore completely unsurprising that rents have risen consistently over that period.

Given that the rental sector in Dublin is roughly 50% larger now than a decade ago, it is realistic to think that rents will not start to fall until the number of listings in the capital exceeds 15,000 in a three-month period.

But where will these extra 2,500 rental homes every month come from? They won’t come from the stock of owner-occupied homes – those are also in short enough supply. There are some quick wins dotted around the capital in the form of vacant homes and there is also potential for over-the-shop conversions.

But in each case, this will take time and the potential should not be overstated. Even the most optimistic estimates of over-the-shop space puts the total number of new homes at perhaps 5,000 or 6,000 – roughly two months supply. Similarly, even a halving of the vacancy rate in County Dublin would only bring a few months on to the market.

Obviously, either or both of these would certainly be welcome. But there are reasons that these haven’t happened already, despite an almost-doubling of rents in the capital. The poor utilization of our built stock stems from a heady cocktail that includes – but is not limited to – low property taxes, incomplete registry of title, conflicting regulations and weak local authorities.

Tackling any one of those issues would take years, let alone all. Thus, while local authorities and the national policymakers should not ignore vacancy, the real solution lies in building new homes. With an average lease length of roughly 3 years, this means that the city needs about 800 homes a month – or an apartment block of 200 homes opening every week.

With only a tiny fraction of that currently being built, the challenge for policymakers is to transform the city’s construction sector into one of the best in the world at building apartments.

==

An edited version of this post was originally published in my column in the Sunday Independent.

Is David McWilliams right – are we in another housing bubble?

The current state of Ireland’s housing system is well known: plagued by a chronic and growing shortage, especially in Dublin, both sale and rental prices increase quarter on quarter. Since the third quarter of 2012, sale prices in the capital have increased in all but three quarters. In the rental market, they have increased in every single quarter.

The fact that both sale and rental prices have now been rising longer than they were falling has led some to argue that we are in another housing bubble. Among the most prominent of those arguing another bubble is coming is David McWilliams.

On the face of it, one would have to be brave to bet against him. David argued a housing bubble was coming down the tracks as early as the late 1990s. Some write this off as “a stopped clock is right twice a day”. However, watching on YouTube his debate with Austin Hughes in October 2003, it is not the prediction of a crash that is arresting, rather how eerily accurate the mechanisms and fallout.

A couple of weeks ago, David argued that he believes there is another bubble building ‘very rapidly’ and that a crash will happen in ‘the coming years’. Outlining his case, he made two main points. The first is that you don’t need credit to have a housing bubble. The second concerns the price of a home relative to incomes. David, and others, believe that is simply not sustainable for three-bed semis in Dublin to cost €450,000 when an average wage is one tenth of this.

There are two minor quibbles with this latter point. The first is that almost nobody buys a property on their own anymore. The average new mortgage has gone from having 1.3 incomes in the 1990s to 1.7 incomes now. So the proper income in David’s example would be the average household income of roughly €75,000, not a single earner.

Also, while appealing, the average property price is not the correct one to use in this case. First-time buyers don’t buy the average property. They typically buy newly built homes at the edge of the city, in other words far cheaper than the overall average. So instead of €450,000 to €45,000, the real comparison is probably more likely €325,000 to €75,000.

Still, that means that the house price is 4.3 times income, well above the typical level regarded as affordable, which is three times income. But t is important to distinguish between what is healthy and what is sustainable. It is not healthy to have high housing prices in major cities, compared to people’s incomes. But cities around the world have been living with this for close to fifty years in some cases.

In Ireland, the ‘Dublin premium’ is just 30 years old. Up until that point, the average price of a home in the capital was the same as the rest of the country. It probably had a bedroom less and a much smaller garden but the price was effectively the same.

But this point works both ways. This is a phenomenon that has been building up for 30 years. It stems from restrictions on land use – the hidden costs of planning and zoning – that prevent housing supply from meeting housing demand. Other cities have had similar supply shortages for 50 years, so it is not obvious to me that expensive housing per se is enough to cause a collapse.

David’s other point is that you don’t need credit for a housing bubble. This is presumably to pre-empt someone like me arguing that the new Central Bank mortgage rules effectively rule out the type of lending that cause the 2000s bubble and subsequent crash.

The godfather of studying bubbles, Charles Kindleberger, argued that a rush of capital (i.e. money) was a necessary ingredient for any bubble. Capital comes in two forms: credit or equity, in other words borrowing to buy or using up savings to buy.

There are plenty of examples of bubbles where people have cashed in their savings to buy at unsustainable prices. It is argued that the dot-com bubble is the best recent example of this. What makes David’s argument trickier to carry over to housing is that real estate is typically very highly leverage in the first place; in other words, there’s lots of debt associated with property.

A savings-fuelled housing bubble is a much rarer phenomenon. Nonetheless, Australia, New Zealand and Canada have found themselves worrying about this problem as their housing markets bear the brunt of Chinese savers’ desire for external assets. But it’s not just about ownership. The key thing about a bubble is the ratio of sale to rental prices.

In markets like Auckland, Sydney and Vancouver, foreign purchasers of property left their dwellings empty, effectively taking them out of the market. But if something should trigger a departure for these buyers, a flood of new homes would come on to the market.

Earlier, I mentioned that both sale and rental prices have been increasing. Indeed, since 2012, rental prices have actually increased by more, in percentage terms, than sale prices. Rental prices are up three quarters in Dublin and by half outside Dublin from their lowest point. Sale prices are up by 60% in the capital and 47% elsewhere.

The single best barometer of a housing bubble remains the ratio of sale to rental prices in housing. It is generally thought of as safe to pay 20-25 times the annual rent to buy a home. This is the same a home giving you a return of 4% to 5% a year. Those buying four-bedroom homes in West Dublin currently are spending 23 times the rent on average, right in the safe zone.

In the Celtic Tiger bubble, people were prepared to pay 40-50 times and in some cases up to 100 times the annual rent. That key difference tells me that we should not be fighting the last war, when it comes to housing. We have a shortage of housing and that gets tackled one way only: more homes.

==

An edited version of this post was originally published in my column in the Sunday Independent.