Unlocking the Publisher’s Digest: Insights and Concerns with UK Labor Market Data
Title: Unmasking the Reliability of UK Labor Market Data: Insights and Concerns
Introduction:
In this fast-paced digital age, accurate and reliable data is crucial for making informed decisions, particularly when it comes to economic indicators like the labor market. However, recent concerns have been raised about the quality and reliability of labor market data in the UK. This article will delve into the intricacies of the issue, exploring the challenges faced by the Office for National Statistics (ONS) and the potential impact on the Bank of England’s decision-making process.
The Decline in Response Rates: A Long-standing Challenge
One of the key issues affecting the reliability of labor market data is the decline in response rates for surveys conducted by the ONS. This decline has been a long-standing concern, but it has become particularly pronounced during the pandemic. The shift from face-to-face data collection to telephone interviews resulted in a dramatic drop in response rates, making it difficult to capture a representative sample of the population.
Furthermore, the decline in response rates has skewed the data towards older and inactive individuals, potentially underrepresenting the experiences of the active working-age population. This presents a significant challenge, as it becomes increasingly clear that the survey respondents may not be an accurate reflection of the overall labor market.
Alternative Data Sources: A Potential Solution?
Given the uncertainties surrounding the reliability of the ONS Labor Force Survey (LFS), it becomes crucial to explore alternative data sources that could provide a more accurate picture of the labor market. One potential candidate is average weekly earnings (AWE) data, which is collected from trustworthy companies rather than relying on individual responses.
Unlike the LFS, AWE data has experienced a stable response rate of around 80%, with only a slight decline during the pandemic. This suggests that AWE data could offer more reliable insights into wage growth trends. However, it is essential to acknowledge that AWE data has limitations, such as its exclusion of companies with fewer than 20 employees and variations in response rates across different industries.
Challenges with Other Candidates
While AWE data shows promise, it is not without its caveats. Other potential candidates such as the Recruitment and Employment Confederation (REC) pay indicator and effective tracking wages face their own challenges. The REC pay indicator provides valuable insights into pay growth but focuses solely on new hires, limiting its scope. Effective tracking wages, although offering early signals, is subject to challenges in data interpretation.
Additionally, the measurement of remuneration through pay slips, while population-based, is often considered experimental data and subject to frequent revisions. Lastly, the LFS wage growth measure, which relies on employee-reported data, suffers from limited response rates and sample sizes, hindering its reliability.
Implications for the Bank of England and Monetary Policy
With doubts surrounding the reliability of labor market data, the Bank of England’s decision-making process may face challenges. The uncertainty in data collection and potential biases could lead to a cautious approach by the Monetary Policy Committee (MPC). The economic literature suggests that caution is advisable in the face of uncertainty, as errors due to uncertainty can be amplified.
This cautious approach by the MPC could impact the future trajectory of interest rates and other monetary policy decisions. If the reliability of unemployment data is questioned, the bank may pay less attention to it, potentially keeping rates higher for longer. Conversely, doubts about the official wages data might lead to less aggressive reactions to strong data in the coming months.
Conclusion:
The reliability of labor market data in the UK is a pressing issue that demands attention. The decline in response rates and potential biases in survey samples pose significant challenges for the ONS and the Bank of England. While alternative data sources like average weekly earnings show promise, they are not without limitations.
Moving forward, it is crucial for policymakers to take a cautious approach, considering the uncertainty surrounding data collection techniques. Exploring multiple sources and integrating data from different perspectives will be essential to gain a holistic understanding of the labor market. By doing so, policymakers can make more informed decisions that accurately reflect the state of the economy and support sustainable growth.
Summary:
The reliability of labor market data in the UK has come into question due to declining response rates and potential biases in survey samples. The Office for National Statistics (ONS) has implemented measures to address these issues, but concerns remain. As a result, alternative data sources like average weekly earnings (AWE) are being considered for a more accurate assessment of the labor market. However, other candidates present their own challenges. The Bank of England’s Monetary Policy Committee (MPC) may adopt a cautious approach in light of data uncertainties, impacting future policy decisions. Overall, policymakers must navigate these challenges to ensure reliable and representative labor market insights for informed decision-making.
—————————————————-
| Article | Link |
|---|---|
| UK Artful Impressions | Premiere Etsy Store |
| Sponsored Content | View |
| 90’s Rock Band Review | View |
| Ted Lasso’s MacBook Guide | View |
| Nature’s Secret to More Energy | View |
| Ancient Recipe for Weight Loss | View |
| MacBook Air i3 vs i5 | View |
| You Need a VPN in 2023 – Liberty Shield | View |
Unlock the Publisher’s Digest for free
Roula Khalaf, editor of the FT, selects her favorite stories in this weekly newsletter.
In by the Office for National Statistics while Alphaville simply tries to enjoy a succulent meal (emphasis ours):
To improve the quality of our labor market estimates, including response rates, we are transforming the Labor Force Survey by asking more people in different ways about their labor market engagement. The first data from this new transformed survey will be published in our main releases starting next spring.
As we have explained previously – and in common with other national statistical offices – response rates for some key ONS household surveys are falling. This has complicated the production of some indicators based on the current UK Labor Force Survey (LFS).
For this reason, the publication of the next ONS Labor Force Survey (previously due on Tuesday 17 October) has been slightly postponed to give us more time to produce the best possible estimates of the labor market using the best data sources available. This data will now be published on Tuesday 24 October.
However, HMRC earnings data, vacancies and real-time information will still be published as originally scheduled on Tuesday 17 October.
Womp womp.
As we wrote in May, the decline in LFS response rates has raised questions about the quality of the data (which is obviously a rather crucial concern when the monetary policy situation is so delicately balanced). The pandemic exacerbated the problems, but things got worse a long, long time. Here’s a reminder of the ugly term trend (which has only gotten worse since the limit used here):
A note from Nomura this morning makes for timely additional reading. Analysts wrote:
The reason for these questions is that the survey response rate dropped dramatically at the start of the pandemic, in part because face-to-face data collection was suspended and replaced with telephone interviews. In July 2020, the initial sample size was doubled in an attempt to compensate for the lower response rate, but as of July 2023 the achieved sample size is still half the 2013 level. Even before Covid, many other developed countries were to conduct larger samples than in the UK in order to produce official labor market estimates. The response rate is just a quarter of the 2011 level, standing at just 14.6% in the second quarter of 2023 (see Fig. 1). This is a trend that we have observed to some extent internationally, with the response rate for non-agricultural wage surveys also declining substantially.
In November 2022, the ONS launched a ‘knock-to-nudge’ program to encourage responses from families in under-represented areas using face-to-face engagement on their doors. This should have made the results more accurate, but it is possible that repeated methodological breaks in sample collection made the data more difficult to interpret/compare.
Another problem is that the sample is predominantly skewed towards older and inactive people. Therefore, we believe that the investigation does not focus on the most important subset (those of “active” age). There may also be a bias in the survey among the majority of older workers (i.e. the survey may be more accurate for older age groups than for younger age groups).
This is perhaps a roundabout way of saying that older people pick up the phone when a random number rings, while younger people do not. Of course, it is undesirable to conduct an employment survey in which the respondents are disproportionately older and inactive.
If the LFS ceases to be reliable, what other data can the Bank of England turn to?
Average weekly earnings are a strong candidate. Unlike the LFS, Nomura points out, its response rate – collected from trustworthy companies rather than untrustworthy people – is much better, remaining stable at 80% after only a slight decline during the pandemic. But it’s not perfect, as Nomura notes:
There are still some caveats to consider. First, the AWE data doesn’t look at companies with fewer than 20 employees. To the extent that wage growth is different across these companies, because they are start-ups or less profitable small businesses that cannot afford to give large wage increases, this could introduce a bias into the reported data. Second, the response rate for AWE varies across industries. It is highest for managerial and professional firms (85%) and lowest for routine and unskilled industries (75%). Again this could mean that the signal is more reliable for sectors with substantially different wage growth.
There are similar problems with the other likely candidates (italics are):
THE REC pay indicator it was also mentioned by the BoE. It’s useful but it’s a flow (while AWE is a stock), so only consider pay growth for new hires. THE Effectively tracking wages has similar problems; however, flow data can provide earlier signals than inventory data. There are also challenges in interpreting other salary data. THE pay slips, measurement of remuneration it is rarely discussed by the BoE, probably because it is classified by the ONS as “experimental data” and subject to frequent and extensive revisions (however these data are at least population-based rather than sample-based). There is also the LFS wage growth measure which asks employees, not companies, for their pay. But, again, this suffers from a limited response rate and sample size.
Nomura’s conclusion is that this uncertainty should be a cause for caution on the part of the Monetary Policy Committee:
The economic literature suggests that it is better for central banks to be cautious in the face of uncertainty (see two papers from BoE AND ECB). This is especially true when the uncertainty is multiplicative. In the case of labor markets, if there are doubts about the output gap this suggests that there will be doubts about the slope of the Phillips curve. In this scenario, Brainard (1967) proposed central banks should act with caution because any errors due to uncertainty could be amplified. Intuitively, this makes sense: if a signal is less reliable, you react less to what it tells you.
At the moment, unemployment data points to the possibility of a substantial slowdown emerging in the UK economy if the current rise continues. If the Bank questions the reliability of this data due to collection problems, then there is a risk that it will pay less attention to it and potentially keep rates higher for longer than necessary (especially if the MPC ends up doubt what turns out to be a real and persistent increase in unemployment). On the other hand, if there are doubts about the strength of official wages data, the BoE may be less inclined to react aggressively to any strong data in the coming months. In general, the BoE may start to give more weight to data perceived as more reliable, such as consumer prices.
Further reading
— The LFS manpower has been lost
— Can we really trust Britain’s economic data? (The Telegraph)
—————————————————-