“Figures often beguile me, particularly when I have the arranging of them myself; in which case the remark attributed to Disraeli would often apply with justice and force: There are three kinds of lies: lies, damned lies, and statistics”, wrote Mark Twain in Chapters of my Autobiography, a voluminous collection of musings, anecdotes and reminiscences he published in the literary journal North American Review in his last years.
Twain’s musing helped popularise the now well worn phrase “lies, damned lies and statistics”, which he erroneously attributed to Disraeli.
The origin of the phrase remains a mystery.
Twain is of course not alone in being beguiled by figures. It is an antipathy shared by a considerable majority of people.
It has been exploited throughout history to fool people, shareholders and citizens particularly, about the true state of their financial affairs for years, even decades, with fairly simple sleights of hand.
In reality, numbers do not lie. It is people who try and lie with them. And lying with numbers is in fact not sustainable.
It relies on the Twain antipathy — the knowledge that most people are loath to spend hours poring over them. Also you can cook numbers, but not the reality.
Sooner or later, the reality will come and hit you on the head.
The Greek Government cooked its books to get into the Eurozone and continued to do so to meet the “stability pact” targets— its chief statistician was nicknamed “magician” on account of how he could make inflation, deficits and losses to vanish.
The government recently issued a seemingly official statement validating the figure of Sh568 billion that the President cited in his statement on the teachers' strike.
When The National Treasury issues an official statement with a detailed breakdown of figures, many people are inclined to take that as the last word on the subject.
One of my statistics teachers used to say that if you torture data long enough, they will confess. Let’s see.
The Sh568 billion summarises as follows: teachers salaries Sh141 billion, national government salaries (including parliament, judiciary and constitutional bodies) Sh94 billion, allowances of Sh66 billion (including Sh9.6 billion for temporary staff), parastatals and military Sh153 billion, and counties Sh107 billion.
WHERE IS THE HARMONY
To arrive at the actual wage bill requires adding up salaries and allowances.
The only allowances itemised for teachers is medical allowances of Sh4.8 billion which brings the total teachers wage bill to Sh146 billion.
There is an item for hardship allowance of Sh8 billion but it is not stated whether this includes hardship allowances for teachers as well.
If it does, I could be understating the teachers total wage bill, but not by much. Adding up the salaries and allowances, the wage bill of the rest of national government comes to Sh145 billion about the same as that of teachers.
An anomaly is readily apparent. The Treasury’s statement, and indeed the budget, does not include the number of people on the public payroll, but these figures are published by the Kenya National Bureau of Statistics (KNBS) in the annual Economic Survey.
The latest published figures are 280,000 teachers and 180,000 people on the national government payroll.
If their wage bill is about the same, then there is a very substantial wage differential between teachers and the rest of national government.
The average monthly pay for teachers works out to Sh43,000 and that of the rest of national government to Sh67,000 a month, that’s a difference of Sh23,000 a month, or 60 per cent more.
MIDDLE JOB GROUPS
One of the government’s arguments in the teachers pay dispute is that their salary has been harmonised with that of the rest of the government. Where is the harmony?
We know that teachers are bunched in the middle job groups, while the mainstream civil service is bottom heavy, therefore the average salary for teachers should be higher than that of the mainstream civil service. What is the source of the anomaly?
The basic salary bill for the core national public service, excluding police and prison staff is given as Sh26 billion.
Compare that to “Other allowances” amounted to Sh26.9 billion. We are not told who gets these allowances, but clearly, the figures are revealing.
The examples given are leave and transfer but these two cannot possibly be that much, unless most civil servants are transferred every month.
It is doubtful that much of these allowances are earned by junior civil servants. We have recently learned from an audit report that a handful of Communications Authority executives made Sh300 million from travel allowances alone.
It is unlikely that this racket is peculiar to the Communications Authority. This is one of the dirty secrets that explains the game of smoke and mirrors we are treated to on matters of the wage bill and public wage policy.
Time after time, recommendations have been made to rationalise and consolidate these allowances into salary, including a report that former Finance PS Harris Mule, Prof Terry Ryan and I wrote for the government in the wake of the last teachers pay crisis a decade ago.
The Salaries and Remuneration Commission has been making the same noises. Why is this resisted? Simple.
If pay is rationalised and made transparent, it will make it difficult for the top civil servants to pay themselves “undercover”.
One of the sure fire smoke screens that is guaranteed to divert public attention is the cost of the new constitutional offices.
Their total wage bill last year was Sh4 billion. The top bureaucrats drew Sh4.7 billion in “extraneous” and “responsibility” allowances — which by the way are not part of the Sh29.6 billion “Other allowances” line item.
If you remove the genuine allowances that every civil servant is entitled to, we are looking at discretionally allowances in the order of Sh20 billion or more that were drawn by top public officials probably no more than 5 per cent of the civil service.
They have, in their bank accounts, millions of reasons to fight rational public pay policy, millions of reasons to massage wage bill, millions of reasons to fan fiscal hysteria whenever ordinary public servants demand better pay.
Next to teachers, the Treasury would have us believe that parastatals have the biggest wage outlay after teachers.
But there are two problems with the parastatal wage numbers. First, they have combined parastatals with military.
This we can assume is to hide the actual military strength which I suppose is a state secret. But why hide it in the parastatal wage bill?
It seems to me imminently more sensible to give one aggregate figure for the national civil and uniformed services.
The second problem is that the number is in fact not a wage bill, but a percentage of the transfers from the government that the Treasury estimates to be the wage bill.
Even this we can’t be sure as the explanatory note puts this at between 60 and 70 per cent. This is very significant.
Getting the actual ratio ought to be a simple matter of perusing the financial statements of the parastatals. It is not.
I have been unable to find a single financial statement of the big infrastructure parastatals that receive budget transfers—Geothermal Development Company, the Rural Electrification Authority the Urban and Rural Roads Authorities) and such as I could find e.g. the Kenya Roads Board do not provide that information in their audited accounts — the published accounts are basically useless.
It is conceivable, but inexcusable, that the Treasury does not have the information either hence the crude 60 to 70 per cent estimate.
But this percentage is implausible. Last year, the Ministry of Energy budget shows Sh79 billion for power generation transmission and distribution.
The ministry does not do these things so we can safely presume that this money is transferred to the relevant parastatals.
The Treasury’s assumption puts the wage bill of these three or four parastatals at between Sh47 billion and Sh55 billion — double the salary wage bill of the core civil service!
Finally, the Treasury’s statement gives a wage to expenditure ratio of 52 per cent.
Even if the wage bill was correct, this ratio would still be misleading.
The revenue figure used in the ratio is the national government’s ordinary revenue which includes national government tax revenue and other recurrent revenues, such as fees and levies collected by ministries and departments, referred to in budget jargon as Appropriations-in-Aid (A-in -A).
It does not include revenues of parastatals and county governments, that is, it takes an inflated consolidated public sector wage bill, expressed as a percentage of central government revenues.
The result is that the ratio is inflated considerably. This is straightforward mischief intended for the consumption of masses beguiled by statistics.
The correct ratio should be either consolidated public sector wage as percentage of consolidated revenues or central government wage bill as percentage of central government revenues.
With such plentiful room for statistical manoeuvre, the public wage bill can be any number that the mandarins want it to be.