The Property Tax Calculator – how does it work?

The Daft Property Tax Calculator

Last week we launched a Property Tax Calculator on It’s a simple tool which asks for information on a property’s location, bedrooms, bathrooms and type, before giving an estimated value for the property. This in turn calculates the Local Property Tax (LPT) due to be paid when the new scheme starts in July. So, where did the idea come from, and how does it work?

Where did this come from?

One of the conditions of the bailout the Irish government received from the IMF was the introduction of a residential property tax. The tax paid is to be based off a self-assessment of the property’s value. This tax is due to be levied in July this year for the last 6 months of 2013, then from January 2014 be billed annually to cover the full 12 months of the year.

In early March, the Revenue Commissioners released an online tool which aims to help people estimate the value of their property. The Revenue have been at pains to stress that this tool is not definitive, and should be taken only as a starting point for valuation. There was much debate about the accuracy of this tool, due to the limited amount of data it has for each property (area and rough property age). This leads to average prices being shown for an area which may have, for example, both 1 bed apartments and 4 bed semi-detached houses. The value of each of these properties would be wildly different, but the revenue tool shows them both in the same tax band. Not ideal, and it’s caused a lot of confusion for people worried about declaring themselves in the wrong band. The Revenue have publicly acknowledged that up to 50% of properties may be valued in the wrong tax band initially. Potentially a fairly big margin of error there.

One of the big issues the Revenue have is that they just don’t have the data quality needed to carry out this sort of exercise. They may know that a property exists in a given location, but don’t have access to it’s size, type, whether it’s been extended, etc, so they’re starting at a disadvantage here. All of this information exists somewhere within various government departments, but a lack of joined-up data is a recurring theme in Irish government. That’s a rant for another day!

Is there another way to do it?

So, if the Revenue don’t have great data, what can be done? On, we regularly release property market analysis via the Daft Report. These reports are based on the details users enter when placing or editing an ad, so is much richer than the revenue data. We have access to things like bedroom number and property type, which can have a big impact on a property’s value. Daft’s economist Ronan Lyons got to work on a model similar to that behind the Daft Report which would allow us to use the rich data we already have from property listings to hopefully give a more accurate valuation tool.

Ronan split the country in to 5 large regions (Dublin, Other Cities, Leinster, Munster and Connaught-Ulster), as well as in to 365 smaller areas (micro-markets), which are generated by taking the areas offered in the Daft search and grouping them by proximity and similarity. Each of these micro-markets was given a coefficient so that we could say that all else being equal, a property in micro-market A is worth 20% more than micro-market B.

Coefficients were then generated for property features (type, number of beds, baths, presence of a garden) based on the property’s location. Once the hard work of evaluating all of these coefficients is done, they get multiplied together to work out a property’s price relative to a control, base property, and this give’s the property’s approximate current value. One of the drawbacks of using Daft data is that people list the asking prices they hope to achieve on the site, which isn’t always the same as a sale price. Based on 2012 analysis, a 10% discount is applied to allow for the drop between asking and realised prices.

Ronan goes in to a good bit more detail on the ins and outs of how these coefficients were generated in a great post on his site, the summary of it being:

The estimated value of a property is based on an economic hedonic pricing model of all 150,000 listings on the site in 2012 and 2013, which controls for when a property was listed, its location, type and size, and other measurable characteristics.

Does it work?

Initial feedback on the calculator has been very positive, through emails to our support team, feedback on social media and nice write-ups from users, such as this one from FeckTV.

There was also a lengthy discussion on the Joe Duffy radio show (Ireland’s most popular phone-in radio show) about the accuracy level of the Daft estimations versus those of the revenue. As an aside, I took a quick look at the traffic impact Joe Duffy has when mentioning a site.

The model is still not perfect – it does not allow for things like the size of the property or improvements made such as a conservatory or good insulation (high energy rating). What we hope that it does do is provide a little more useful information for home-owners ahead of the filing deadline at the end of May.

Share This Article

Related Articles

A/B testing headlines on social media

A/B testing is a great way to determine the most compelling headline for an article, but is made difficult by the way social media networks cache link previews. What's the secret to effectively running A/B headline tests, and optimising engagement rates?

Lazy loading background images to improve load time performance

Lazy loading of images helps to radically speed up initial page load. Rich site designs often call for background images, which can't be lazily loaded in the same way. How can we keep our designs, while optimising for a fast initial load?

Font Subsetting - shrink down font files to speed up page loads

Fonts are one of the largest resources on any page after images, and can have a big impact on CLS when they vary in size from the underlying system font. Font subsetting allows us to radically shrink font file sizes, speed up initial page loads, and improve our page speed scores.

Calculating rolling averages with Laravel Collections

Rolling averages are perfect for smoothing out time-series data, helping you to gain insight from noisy graphs and tables. This new package adds first-class support to Laravel Collections for rolling average calculation.

Minimising Cumulative Layout Shift (CLS) When Loading Responsive Ads

Responsive ads are a great way to maximise publisher revenue from display ads. Not knowing the size of the ad to be served in advance can have a big impact on Cumulative Layout Shift (CLS), and, ultimately, Google rankings. How do we maximise revenue while minimising CLS impact?