What does it mean to value land fairly and credibly in practice?

Guest blog: Dr Bryonny Goodwin-Hawkins, UKRI Policy Fellow

Some kinds of land reform – especially those delivered through the tax system – rely on valuing land. Land valuation may seem like a technical matter of data and professional expertise, but it’s deceptively challenging in practice, and that matters for reform. At a recent conference on land valuation in Wales, I was invited to reflect on valuation from a policy perspective. While the discussion focused on Wales, many of the issues raised will be familiar to policymakers across the UK, including in Scotland, where land value also sits at the heart of debates about tax and land reform.

What does “land value” mean?

Land is fixed in place and finite in supply. Economists have long recognised that land value reflects differential advantages. For example, quality agricultural land is worth more than windswept, poor soil. Land for housing gains value from nearby amenities, like parks or transport links.

The factors that fundamentally influence land value are hard to individually change. Compost, and you still won’t grow wheat in the uplands. Redecorate, and you still won’t have a sea view or a bus stop outside. This is “unimproved land value” – subtract the “stuff” on-site and the drivers of value are left.

In Scotland, as in Wales, these differences are particularly pronounced, given the mix of dense urban areas, rural settlements, islands, and upland land uses. This diversity makes valuation both important for policy and challenging to carry out consistently.

Why is land valuation difficult?

To know the value of my house, I can go online and look up the sale prices of like properties nearby. I live in a terrace in a popular urban area, so can find many recent, similar sales. If I lived rurally (or in a converted spaceship) I’d have more trouble. Regardless, I won’t find separate prices for the house and land underneath.

This illustrates the key challenges for land valuation:

• We can’t know the market value until property sells, but non-urban markets are slower and less comparable.
• Much urban land is built upon, and we don’t separate land and improvements when we buy or sell.

For most land in Wales, there isn’t a value waiting in a database. We need to approximate the value using indirect data, professional judgment, and many assumptions.

What valuation methods are available?

There is no shortage of ideas about how to value land. During my UKRI Fellowship with the Welsh Government, I identified five types of approach:

• Market-based statistical methods use transaction data to estimate how factors like location and amenities influence values.
• Algorithms and machine-learning are data-intensive and focus on automation and prediction.
• Formula-based approaches, such as simple area-based calculations (£ per m²), are rough and ready.
• Conventional valuation draws on established UK professional practice but requires adaptation to separate land from buildings.
• Innovative or experimental approaches might include participatory methods, self-valuation, multi-criteria tools, and more. 

At the conference, research teams demonstrated these approaches for Wales. No single method won out. Rather, different methods produced plausible results with different strengths and weaknesses.

What matters for policy?

My takeaway from the day is that we can value land in Wales – insofar as we can come up with numbers. Those numbers won’t be perfect (and some could be plain wrong). To find “good enough”, policymakers need to be explicit about choices like: 

• Should valuations prioritise precision, or broad coverage?
• Is transparency more important than technical sophistication?
• How frequently do values need to be updated to remain credible?
• How much accuracy can be traded off against costs?

The point is less to search for a more perfect method than to design a system that is fit for purpose and publicly defensible. Seen this way, land valuation is more than a technical fix – it is a question of governance, legitimacy, credibility, and compromise.