Pay and inflation: examining the link

The link between headline inflation and pay settlements has long been of interest to both wage setters and economic commentators. We review some of the latest developments.

The 2010/11 annual bargaining round is almost upon us, with pay setters now considering their positions in advance of forthcoming salary reviews. In the absence of a yardstick against which to measure the value of any increase, it would be difficult, if not impossible, for wage decision-makers to know where to pitch claims and offers. Fortunately there has been, and continues to be, such a benchmark: headline inflation as denoted by the year-on-year increase in the all-items retail prices index (RPI). This data is produced on a monthly basis by the Office for National Statistics (ONS) (external website).

While the basket of goods used to compile the RPI has changed over the years to more accurately reflect actual purchasing patterns, and the relative weights attached to them, it has remained the standard against which those on both sides of the bargaining table judge the value of a pay award. It is perhaps worth noting that, on the ONS website, included in the list of uses to which the RPI is put is "wage bargaining". This could be interpreted as an official acknowledgment of the central role headline inflation plays in pay setting.

Despite this, various Governments have tried to wean the UK off the RPI in favour of the consumer prices index (CPI). Indeed, this is the current and previous administrations' preferred measure of inflation for macroeconomic purposes. Our 2009 review of pay trends revealed that nearly nine in 10 (86.3%) respondents still refer to the RPI when wage setting. Despite this, our research also indicates that private sector employers have been displaying an increasing interest in the CPI, with almost one-third (32.8%) referring to it when making decisions on pay in the 2009 bargaining round, compared with just 14.7% in 2008.

While the RPI has always been subject to movements both up and down, in recent times there have been some notable fluctuations. For example, over the year to February 2009, headline inflation dipped to nil before entering negative territory (-0.4%) in March 2009. It fell further to -1.6% in June, the lowest figure since the series began in 1948. In the months that followed, the RPI picked up to reach 0.3% in November 2009. Thereafter, there was a sharp acceleration with inflation on the RPI measure reaching 5.3% in April 2010, the highest level for 19 years. Obviously, such swings cause problems for pay decision-makers particularly, but not exclusively, in relation to the formulation of long-term deals as outlined below.

Settlements and inflation

When XpertHR last examined the link between inflation and wage settlements in 2007, we asked whether or not, and to what extent, pay deals are linked to movements in the headline rate of inflation.

In order to provide an answer we plotted the change in RPI against the movement in the XpertHR headline measure of wage settlements (formerly known as the IRS measure of wage settlements), as denoted by the median increase by number of pay reviews, over a 20-year period between April 1986 and March 2006. This analysis was subject to a six-month time lag so that, for example, the figure for the annualised increase in all-items RPI for the 12 months to the end of April 1986 corresponded to the median pay award for October 1986; the May 1986 RPI figure was been plotted against the PABB median for November 1986 and so on.

The six-month time lag was chosen because it is unreasonable to assume that negotiators are in a position to respond immediately to a change in inflation. Neither is it likely that bargainers alter their pay-setting behaviour in response to a single month's movement in the RPI, up or down. It is more likely that, when formulating their positions, negotiators take account of the inflationary climate over the months prior to the anniversary date while, in all probability, also considering inflation forecasts for the period after the implementation date.

It should also be noted that tracking the relationship between pay and inflation is not an exact science. The assumption of a six-month time lag used by XpertHR is something of a "guesstimate" and, as we have pointed out previously it is likely that the correlation between these two variables is actually quite fluid, and may be subject to change over time in response to particular circumstances.

The original graph indicated that, over the two decades covered by the analysis, there has been a striking relationship between wage deals and inflation. However, looking over the period as a whole, while the connection between the XpertHR headline measure of settlements and the RPI appears to be a close one, there are some spells where inflation has outstripped pay settlements (for example from the spring of 1990 until the autumn of 1991) and others where the reverse has been true (for example, from the early months of 2001 until the end of 2002).

The trend continues

We have now extended the period under examination to December 2009 (for headline RPI) and June 2010 (for the XpertHR median pay award). A review of the years from April 2006 to the present day reinforces the view that there is indeed a correlation between inflation and pay deals. In this case, the XpertHR median award has remained consistently above headline RPI for the three years to this year.

Perhaps the most interesting period occurred between the latter half of 2008 and the start of 2009 when the full effects of the global financial crisis were being felt. As inflation dipped below 1%, and then into negative territory, settlement levels eased back, as would be expected, reaching nil in the rolling quarter to the end of July 2009. However, as the period of deflation continued, pay awards recovered from what was a precipitous decline. Indeed, during the months of negative inflation, our headline measure of pay deals, on a six-month lagged basis, never dropped below 1%.

How will settlements respond?

The key question for policy makers is how will pay deals respond to recent movements in RPI, which stood at 4.8% over the year to July 2010? Despite the latest fall in headline RPI, down 0.2 percentage points on the June 2010 figure, history would suggest that settlement levels could pick up as we enter the autumn wage round. However, there are factors that suggest the upward pressure on wages could be mitigated, at least in part:

• First, according to many economic forecasts, headline RPI is predicted to ease back in the latter half of this year and into 2011. It could be the case that, as the most intense months of the annual pay round, January and April, are now behind us, many wage decision-makers (particularly employers) are looking ahead to the new bargaining year and so may take more notice of an anticipated fall in inflation rather than the RPI profile in the recent past. However, trade unions may take a different view as to whether or not past inflation is relevant when pitching their claims, arguing for a "catch-up" increase, particularly where their members had previously been subject to a wage freeze.
• Second, there are lingering concerns about a possible "double-dip" recession, notwithstanding the most recent ONS figures which show that the UK economy grew by 1.2% in the second quarter of 2010 (external website). Indeed, the August 2010 Inflation Report issued by the Bank of England (PDF format 596K) (external website) shows that the Bank has pared back its projections for the likely increase in gross domestic product compared to that made in May 2010 (PDF format 5.53MB) (external website). While this does not mean that the UK economy will tip back into recession, it does indicate that the recovery is still fragile and is set to remain so over the coming months.
• The fact that there was a significant degree of wage flexibility during the last downturn, with employees accepting wage freezes and, in some isolated cases, salary cuts, has been much commented upon. While there is no evidence that the proportion of wage standstills is increasing, the psychological effects on employees of the last, severe, recession may still remain and some workers may be willing to accept below inflation increases in order to preserve jobs. However, this is not universally the case as the results of recent strike ballots at BT and BAA illustrate. Even so, it should be noted that in these instances, while there was a threat of industrial action, it was not actually taken and the settlements, while relatively generous, still yielded increases below headline RPI.
• Wage militancy in the coming months may be confined to certain sectors. For their part, some employers, still facing a degree of economic uncertainty, could respond by offering wage awards at least partly in the form of one-off cash lump sums, rather than the more traditional consolidated percentage increases.
• Further, relatively high unemployment and spare capacity within the economy will also act as a downward pressure on wage settlements, as will the effect of spending cuts on the public sector and those private sector firms heavily reliant on central and local government contracts.

Long-term agreements

There are, of course, other ways in which headline RPI may influence bargaining patterns beyond that of setting the level and evaluating the worth of a standard 12-month pay review. It also has a prominent role to play in the formulation of long-term deals. As we noted in a 2007 analysis the fact that employees are taking a risk when agreeing such deals means it is common for second and, where applicable, subsequent stages of these settlements to include some form of inflation-link, designed to act as a safeguard for both employer and employees.

This can take a number of forms. Arguably the most simple is a formula whereby employees receive a pay rise equivalent to the increase in the movement in prices at a fixed, future, date or dates. For example, this is the basis for the calculation of the second year of a two-year deal covering 350 manual workers at the Kellogg Company's Manchester plant, effective from 1 April 2010. Those covered received a3.7% increase, based on February 2010 headline RPI.

Other deals feature a pay award equivalent to inflation, plus an additional amount, as atSellafield Ltd (formerly part of BNFL) where employees received an overall increase of 4.9%, effective from 1 April this year, with the level of award determined byMarch 2010 headline RPI (4.4%) plus 0.5%.

However, such has been the volatility of headline inflation over the past year or so, even some relatively complex long-term agreements have been unable to cope with unforeseen circumstances. For example, the final year of a three-year deal covering almost 700 workers at Cambridge University Press' Publishing Division should have yielded an increase in line with October 2009 headline RPI from 1 January 2010. However, as this figure was negative (-0.8%), a possibility that, in all probability, would have not entered the minds of those who negotiated the agreement, a 12-month pay freeze was implemented rather than a pay cut. In circumstances such as these, it is hardly surprising that wage setters are wary of long-term agreements.

One way of dealing with such problems is the inclusion of what is known as a "re-opener clause", which allows for negotiations to be resumed should inflation rise above (or fall below) a given level. Those pay setters minded to negotiate new long-term deals despite the current economic climate may turn to such arrangements to insure against unforeseen circumstances.

We set out some examples of current RPI-linked agreements in table 1 below.

Multi-year deals less popular

In light of this uncertainty, it is unsurprising that our January 2010 review of pay trends, found evidence that at the start of the calendar year some employers who had until then tended to favour long-term agreements had reverted, or were likely to revert, to more traditional 12-month reviews.

As the bargaining year has progressed, and more deals have been gathered by our team of specialist pay researchers, it has become apparent that this has indeed been the case. Over the six months to the end of June 2010, 6.8% of reviews on the XpertHR pay databank were a result of a new or existing long-term agreement (defined as a settlement of two years or more in duration). Over the comparable period in 2009, such deals formed 8.7% of all awards.

Despite an overall fall in the proportion of new long-term settlements, there are still new multi-year deals being agreed although by no means all of them have a built-in inflation link. For example, the recent settlement for an estimated 10,000 workers covered by the Scottish and Northern Ireland Joint Industry Board for the Plumbing Industry kicked off with a 12-month pay freeze from 7 June 2010. The second stage of the deal, which comes into effect on 6 June 2011, will yield a 2% basic pay rise. The agreement does not feature a link to inflation.

Similarly, there is no RPI link in a comprehensive, three-year, pay, job security and industrial relations agreement negotiated between the Communication Workers' Union (CWU) and Royal Mail. This features a 2% basic pay rise from 1 April 2010, to be followed by a 1.4% increase in April 2011 and a 3.5% uplift in April 2012.

In contrast, one of the most high-profile of the new crop of inflation-linked deals is that negotiated between the CWU and BT. The agreement includes a 3% basic pay rise from both 1 January 2010 and 2011. The third stage, effective from 1 January 2012, is also worth 3%. However, should November 2011 headline RPI (published in December 2011) be above 3.2% or below 2.5% "then both sides agree to meet and discuss the impact of this, and seek to agree an optimum position. Both parties agree to enter into these discussions to seek to ensure that the impact of a volatile RPI is mitigated, both for employees and the business".

Table 1: Examples of RPI inflation-linked deals