There has been a proliferation of LMI tools coming onto the market recently, notably as UCKES lift the lid on their data through their LMI For All programme, but how does this stack up against EMSI’s data? EMSI have been innovating and driving forward the use of LMI in the UK and globally for over 15 years, investing significant time and effort in building, maintaining and improving our proprietary dataset. But if public data is so freely available, why do we do what we do?
It’s quite simple really. Publicly available datasets have inherent weaknesses, particularly when used in isolation, and these need to be resolved before meaningful decisions can be made. Let’s take a couple of the more common datasets available to us and have a look at these:
NOMIS is the Office for National Statistic’s portal to publicly available official statistics. A large amount of data about the economy and labour market can be accessed here. This data’s strength lies in its detail, both geographically and occupationally/sectorally. That said, its first weakness is that large amounts of the data cannot be published for confidentiality reasons and is suppressed. The second major weakness comes in its timeliness. Most data is already up to two years out of data by the time it is published and so can only really paint a picture of the past, and not the present day or future.
Working Futures is published through UKCES and is one of the three key datasets being made available through LMI For All. This dataset provides employment projections designed to look at how our labour market is likely to change into the future, thus addressing one of the key weaknesses in NOMIS data. That said, it too has some key weaknesses of its own to consider.
Firstly, the level of published detail is not sufficient to make meaningful decisions – exploring where the manufacturing sector across the South East is heading is interesting, but tells us little about which aspects of the sector are driving these trends (manufacturing can be broken into 230 constituent industries from meat processing, to the repair of ships), or how these trends are affecting local economies (the manufacturing profile in Kent and Milton Keynes are very different). The second key weakness is the timeliness of refreshing the forecasts – Working Futures is updated roughly every three years, meaning that the unexpected vagaries of the real world economy render the forecasts quickly out-of-date.
Keeping Strengths; Removing Weaknesses
The challenge to create the most accurate view of the labour market is to harness the strengths of the datasets, but remove as many of the weaknesses as possible. And this is exactly what EMSI’s expert economists do. EMSI have developed a robust methodology that links multiple datasets together to achieve a more accurate solution, harnessing the detail and accuracy of NOMIS data, and using the general direction of travel from Working Futures. We refresh our forecasts annually to take account of changes in local economies that have been documented through updated NOMIS information.
But if public datasets have such weaknesses why do we use these in favour of other information such as job postings? So-called ‘real time LMI’ uses information from published job postings to extract an understanding of the economy. It provides a detailed picture of employment and skills but it too has inherent flaws. You can read a more detailed assessment here, but in summary the main problems are:
- Not all jobs are advertised and so the information gathered is from a limited base
- Some jobs are advertised multiple times and so may be “captured” multiple times
- Seasonal employment trends can tend to provide a skewed picture of the labour market
So although this type of data does a good job in understanding who’s hiring now and what they require, meaningful long term planning decisions cannot be made with any accuracy using this information.
Poor Data = Bad Decisions
While EMSI welcomes the growing awareness of the important role LMI can play in making better local decisions, it is important to understand weaknesses in the data – key decisions made with poor data can be just as costly as decisions made with no data at all. Be mindful of LMI tools that rely heavily on Working Futures data and so carries these weaknesses. Also be aware of job postings based analysis, as they can only ever provide a very limited snapshot of the labour market.
In answer to my initial question, this is why we at EMSI do what we do.
For further information, please contact Andy Durman at firstname.lastname@example.org