Skip to content

Video summary: Microdata perspectives on the UK productivity conundrum - An Update

Released: 04 October 2013

This is a transcript of the video podcast which can be viewed at: youtu.be/hwUHKF_XVEI

Introduction

This is a short video providing an overview of micro-data perspectives on the UK productivity conundrum.

Slide 1: The Productivity Conundrum

Let’s begin by looking at what the productivity conundrum is by using this chart. 

The dotted line shows an index of 100 from quarter 1 2008; so below the line, a measure would be lower than in quarter 1 2008, and above, it would be higher.

We can see that at the beginning of 2008, before the recession, output, shown by the red line, began to fall and, at present, is well below its quarter 1 2008 level.

However at the same time, we can see that hours worked, shown by the green line, has recovered since 2010 and is presently at a level above its pre-recessionary level.

This means that output per hour, or productivity, shown by the blue line, has been falling since quarter 1 2008.

This podcast and the accompanying publication attempts to look at how micro-data can help toward explaining these trends.

Slide 2: Micro-Data

But let’s first explain what micro-data is and how it is compiled. Firms across the country provide information by completing surveys such as the Annual Business Survey, Monthly Business Survey, Community Innovation Survey and many others.

Examining the results of these survey responses at an individual firm level is micro-data analysis. 

This allows us to do two unique things.

Firstly, it enables the analysis of the characteristics of individual firms, and secondly it allows the breakdown of the full sample into groups with similar characteristics.

Let’s look at how micro data differs from macro data. For macro data, we aggregate survey responses so we can look at the input and output by industry and produce estimates of the productivity by industry.

For micro data however, if we look at the firm level responses as our source, we are able to produce productivity by industry and firm.

With this firm level data we can investigate the part played by firm characteristics which influence their productive level.

It must be stressed that the reported results protect in full the confidentiality of individual responses.

Slide 3: Industries and Measures

Now we will take a look at how the micro-data is split and what measures are used to analyse it in a particular example.

This analysis splits the results into three distinct sectors.

MexElec refers to manufacturing firms apart from those in the manufacture of electronic goods.

Secondly, we refer to MServ which includes most market services, though some are not included such as the communications industry.

And finally, when we refer to Elecom we include the two sectors omitted from the previous categories - electronics manufacturing and communication.

These three industries are examined by two measures in the publication.

Firstly, labour productivity which looks at the real gross value added per employed person.

Secondly we have total-factor productivity, which accounts for the changes in total output that are not accounted for by the changes in the quantities of labour and capital, that are used as inputs.


Slide 4: Firm Level Characteristics

We will now look at four of the characteristics highlighted within the analysis, starting with size class.

We will look at the impact of size class in the MexElec industry group on the Labour Productivity Measure.

Size classes are defined by how many employees a firm has - it is split here into four classes as shown.

The chart shows the change in labour productivity for each size class between 2001 and 2010.
Whilst we can see clearly that there is a positive relationship between firm size and productivity it must be noted that the increase in productivity between size class 3 and 4 in the MexElec industry is particularly high.

Moving onto the second characteristic - whether the firm is an exporter or not. We will look at this for the market services industry for both labour productivity and total factor productivity.

The charts show productivity for both exporters and non-exporters as well the total.

A key observation here is that for both measures, productivity of exporting firms in the MServ industry is notably higher than the average.

The third characteristic explores whether the firm is foreign or domestically owned.

For this we look at the labour productivity within the MexElec industry.

A general observation shows how foreign owned firms are more productive than domestic firms in the MexElec industry.

An interesting point is that in the post recession period, productivity of foreign firms has risen sharply in the MexElec industry whilst the productivity of domestic firms has begun a gentle decline.

The final characteristic looks at broadband usage within both the Elecom and MServ industries.

The charts show the labour productivity of firms for which above and below 40% of employees have access to broadband.

The general trend suggests that firms with more than 40% access have been more productive than those with less.

The fall in real productivity that followed the recession for firms with more than 40% access may have been caused by more firms becoming ICT equipped, reducing the productivity advantage. As this is only a sample it could not be determined for certain.

Slide 5: Sample Characteristics

Finally, micro data allows us to explore characteristics of the firms in the sample. Here, we will focus on the distribution of their productivity.

If we take firms in the sample, we can rank them by their level of productivity.

From this we can group them into quartiles and find an average of each quartile. A quartile divides the order into four groups, with each group containing a quarter of the data. We also find the average of the entire sample.

This allows us to form a distribution line as shown. We show the average of the most and least productive quartiles as well as an average of the four quartile averages.

Let’s look at this distribution chart for labour productivity in the MexElec industry.

It can be seen that for MexElec, the distribution has been increasing up to 2008, after which the whole productivity of the industry shifted downward.

END

 

This is a short video from the ONS looking at the UK productivity conundrum from a micro-data perspective.

First it explains what the productivity conundrum is, as well as what micro-data is and how it is obtained.

Next it looks at the different industries and measures that have been used in the accompanying publication to examine the conundrum from a micro-data perspective.

It then goes onto look at the four firm level characteristics used in analysis of the productivity conundrum. These four characteristics are size class, whether a firm is an exporter or not, whether they are foreign owned or not and finally whether they have access to broadband.

Source: Office for National Statistics

Background notes

  1. Details of the policy governing the release of new data are available by visiting www.statisticsauthority.gov.uk/assessment/code-of-practice/index.html or from the Media Relations Office email: media.relations@ons.gsi.gov.uk

Content from the Office for National Statistics.
© Crown Copyright applies unless otherwise stated.