Trouble on the Greek railways (WS)

Summary

More or Less – Trouble on the Greek Railways

BBC Radio 4, May 14, 2012
Host: Tim Harford
Key Guests: Wesley Stevenson, Stefanos Manos, Greg Moisiades, Panos Prevedouros, Dr. Hermann Stern

---

Episode Overview

This episode of More or Less explores two major topics:

1. The oft-repeated claim that Greek railways are so expensive it would be cheaper to send all rail passengers by taxi.
2. The relationship (or lack thereof) between executive pay and company performance across major economies.

Presented with characteristic wit and a forensic approach to numbers, the episode questions received wisdom and provides clear statistical context to both debates.

---

Greek Railways: Debunking the "Cheaper by Taxi" Myth

The Claim and Its Origin

• Stefanos Manos, former Greek finance minister, claimed in 1992 that "it would be cheaper to send [rail passengers] by taxi than by train," a statement that persists in conversations about Greek public sector inefficiency.

  "I decided that it was quite obvious that it would be cheaper to send them there by taxi than by train. And I still feel that it still holds true. In fact, it might be even worse these days." — Stefanos Manos [01:33]

Investigating the Numbers (01:56–03:43)

• Wesley Stevenson investigates whether the claim still holds.

• Greek railways, in 2010, lost around $1.3 billion; using 2007's passenger figure of 1.8 billion kilometers, the cost per passenger-kilometer is about 75 US cents.

• Taxi cost (from Greg Moisiades, a Thessaloniki cab driver):

  • Thessaloniki to Athens: €700 for 520km ($1.55 per km).

  "So that's $1.55 per km, which if I recall correctly is more than twice as much as it costs to send people by train. And so Manos was wrong." — Tim Harford [03:34]

Wider Context & Real Issue (04:04–05:38)

• Panos Prevedouros, transport engineering professor, explains that rail’s market share is negligible (long-distance rail at 2.7% of passengers; freight at only 0.08%). Over $13 billion was invested over 16 years with little result.

  "In terms of freight it’s truly a joke really because it’s 0.08% of the freight. So basically it carries no freight to speak of." — Panos Prevedouros [04:21]

• Other alternatives:

  • Bus: €32 one-way (10 cents/km)
  • Airplane: €75 (18 cents/km)
  • Private car: €95 (23 cents/km)

• Train is 75 cents/km—more expensive than all alternatives (except taxi).

  "All this is comparing with 75 cents a kilometre being paid to subsidise the trains." — Tim Harford [05:11]

• The core issue: vast debts ($10bn in 2010, $520m in interest annually) guaranteed by the Greek government.

  "So even if the Greeks closed the system today and told everybody to use coaches, the Greek government would still be on the hook for all these huge debts." — Wesley Stevenson [06:08]

---

Executive Pay vs. Performance: Analyzing the Link

Introduction (06:08–07:09)

• Moves to a statistical analysis of CEO pay and company performance across the US, UK, Switzerland, and Germany, referencing shareholder anger at high executive compensation.

  “Pay peanuts, get monkeys. What’s infuriating shareholders is they seem to be paying millions of dollars and getting monkeys.” — Tim Harford [06:14]

Findings from Dr. Hermann Stern of Obamat (07:09–08:51)

• Study analyzed top 100 S&P companies (US), using operating cash flow and total shareholder return versus CEO pay.

  "I found absolutely no connection between pay and performance. If there was a connection on a scale from 1 to 10, the connection is less than 1." — Dr. Hermann Stern [07:13]

• Same pattern found in UK, Switzerland. Germany showed some correlation, attributed to tighter legal controls on board pay.

  "Germany has a special situation because they have a much tighter legal framework that holds boards much more accountable for their pay policies." — Dr. Hermann Stern [07:42]

Possible Defense for Executives

• Tim suggests that CEO pay might look inflated due to past stock options and market cycles.

  "If I'm a chief executive... and I sit on these share options... I'm going to look as though I was paid a huge amount for not doing very much." — Tim Harford [07:56]

• Stern counters that most compensation reflects overall market uplift, not individual outperformance.

  “A lot of those compensation components are absolute compensation, which means they reward not just the performance of that company, but actually also the wider economic cycles." — Dr. Hermann Stern [08:24] “So you’re rewarding a mining company because the price of copper is high. You’re rewarding an oil company because the price of oil is high, whereas that’s not really a measure of how effectively the company is being managed.” — Tim Harford [08:40] "Exactly. This is one source for a lack of correlation." — Dr. Hermann Stern [08:51]

---

Notable Quotes & Memorable Moments

• Stefanos Manos (re: Greek railways):
  “I still feel that it still holds true. In fact, it might be even worse these days.” [01:33]
• Greg Moisiades (on taxi fare):
  “It costs €700... It is 520 kilometers. And it takes about five hours.” [03:16]
• Tim Harford (debunking the myth):
  “So Manus was wrong.” [03:34]
• Panos Prevedouros (on rail inefficiency):
  “In terms of freight, it’s truly a joke really because it’s 0.08% of the freight.” [04:21]
• Dr. Hermann Stern (on executive pay):
  "I found absolutely no connection between pay and performance." [07:13]

---

Key Takeaways

• Greek railways are indeed inefficient and heavily subsidized, but the “cheaper by taxi” claim no longer holds true. Rail is, however, much more expensive than bus, car, or even air travel.
• The real problem for Greece is the railway’s vast debts—not just operating losses.
• Across the US, UK, and Switzerland, there is no meaningful correlation between what chief executives are paid and how their companies perform. Germany is a partial exception, suggesting tighter regulation may improve alignment.

---

Timestamps for Important Segments

• [01:03] – Origin of “cheaper by taxi” claim
• [01:56] – Difficulty verifying 1992 data; 2010 financials
• [03:09] – Taxi fare and distance calculation
• [03:34] – “Cheaper by taxi” claim debunked
• [04:21] – Full extent of rail inefficiency; market share stats
• [05:11] – Costs of alternative modes (bus, car, air)
• [05:38] – Explanation of debt problem
• [07:13] – Findings on CEO pay versus company performance
• [08:24] – Flaws in incentive alignment explained

---

For more details or to suggest your own statistical myth for examination, contact More or Less.

Transcript

[A]

Thank you for downloading from the BBC. For details of our complete range of podcasts and our terms of use, go to bbcworldservice.com podcasts this is more or Less your statistical guide to life, the universe and everything. This shorter edition of the programme was broadcast on the BBC World Service. Hello and welcome to More or Less on the BBC World Service, your statistical watchdog to the numbers around you in the news and in life. This week we'll be asking if there's any connection between pay and performance in major corporations. But first, Wesley Stevenson, our very own Greek God, has joined us. What have you got for us this week, Wes?

[B]

Well, in recent editions of More or Less, we've tackled a lot of Eurostats, some of the wad of statistics which have emerged with increasing frequency as the euro crisis has deepened.

[A]

There's something missing there, Wes.

[C]

Hmm.

[A]

That's more like it. Okay. Yes.

[B]

Well, yes, I've been looking at the Greek rail system. For such a small network, there's only one main line. The Greek railways do seem to crop up a lot in the story of how the Greek nation came to be in its current predicament. There's one particular claim which I want to examine, that the Greek railways are so expensive to run, it would be cheaper to send the passengers by taxi. It was first made by Stephanos Manos when he was finance minister in Greece way back in 1992. Mr. Manos assures me that when he said it, at least it was true.

[C]

I knew the number of passengers and I made a very brief estimate of what it would cost to send them from Athens to the north of Greece. And I decided that it was quite obvious that it would be cheaper to send them there by taxi than by train. And I still feel that it still holds true. In fact, it might be even worse. These days.

[B]

It's been hard, impossible actually, to find the figures for 1992 to check this out. But rail experts have told me they think it was probably close to being true in 1992, but a bit of an exaggeration, but Stefanos Manos, the former

[A]

finance minister, he says it's probably still true today.

[B]

Yes, and he's not the only one. So we've been trying to look at the numbers for what's happening now. We know that in 2010, the two parts of the Greek railway were together, making a loss of around $1.3 billion. From that, we can work out how much was being spent on every kilometer each passenger traveled that year. Now, the Greek statistical agency have some useful figures for the number of passenger kilometers traveled each year. And the latest data are from 2007 when passengers traveled 1.8 billion kilometers. Now if we assume that's roughly the same for 2010, that means it's costing the Greek Railways 75 US cents for every kilometer each of its passengers traveled, which is quite a lot.

[A]

Ok, we just need to know now what it costs to travel around by taxi in Greece.

[B]

Yes, and Greg Moisiades could tell me that. He's a cab driver. Mr. Taxi in Thessaloniki. Oh, hello, is that Greg?

[C]

Yes, Greg speaking.

[B]

Can you tell me how much it would be to take a taxi from Thessaloniki to Athens?

[C]

It costs €700.

[B]

€700. And how far is it, do you know?

[C]

It is 520 kilometers. And it takes about five hours to get from Thessaloniki to Athens on a motorway.

[B]

Okay, that's brilliant. Well, thank you very much for that.

[C]

Thank you.

[A]

So that's $1 55 per km, which if I recall correctly is more than twice as much as it costs to send people by train. And so Manus was wrong.

[B]

This looks like another convenient story that fits the narrative of Greek waste. But like so many others, it just isn't true of.

[A]

Okay, very good. But there is a bigger picture here and we might be in danger of missing it. Mr. Manos, headline grabbing statement might have been an exaggeration, but there was a fundamental truth here, which is the Greek railway is rather expensive to operate and it's in a bit of a mess, isn't it?

[B]

Yes, I think that's right. And Panos Prevedoros, a professor of transport engineering at the University of Hawaii.

[A]

Hang on, hang on. You've been discussing the Greek rail network with a bloke in Hawaii.

[B]

Well yes, of course. He's an expert in this sort of thing and he paints a pretty bleak picture of the Greek railway.

[D]

Over $13 billion has been pumped in in the last 15, 16 years. In terms of passenger, long distance rail has 2.7% of the share. And in terms of freight, it's truly a joke really because it's 0.08% of the freight. So basically it carries no freight to speak of.

[B]

Panels points out there are other good alternatives to the train other than taxis.

[D]

For the main trunk trip from Athens to Thessaloniki, from essentially the center to the north, the bus charges €32 one way trip. If you do it by air, the rough approximation is 75 Euro. And if you were to do the same trip by a mid sized car, that would be roughly €95. So there are for competitors in the mix.

[A]

So just to run through that, in US dollars, it would cost 10 cents a kilometer to send people by bus, 18 cents a kilometer to send them by plane, 23 cents a kilometer to send them by private car. And all this is comparing with 75 cents a kilometre being paid to subsidise

[B]

the trains, not to mention the money that could be made selling off the potentially more profitable routes, the rolling stock and the other assets.

[A]

But isn't the elephant in the room the fact that the Greek railway itself has huge debts?

[B]

Yes, and in the past, the Greek government acted as a guarantor for the money the railway borrowed, which is something other European states do. But the difference with the Greek railways is it had a turnover in the low hundreds of millions, but it was able to borrow in the billions. And in 2010, its debt was $10 billion and cost $520 million in interest payments. So even if the Greeks closed the system today and told everybody to use coaches, the Greek government would still be on the hook for all these huge debts.

[A]

Thank you, Wes. Now, it's been quite a month for shareholder rebellions. It's not that shareholders necessarily mind paying their chief executives well. After all, there is a saying, pay peanuts, get monkeys. What's infuriating shareholders is they seem to be paying millions of dollars and getting monkeys. Anyway, all this got us thinking about the link between pay and performance. Are shareholders getting value for money from their highly paid chief executives? A Swiss financial research company called Obamat has been analyzing the pay and performance of the top companies in the United States, Germany, Switzerland, and the United Kingdom. I spoke to Dr. Herman Stern from Obamat and began by asking him what it found in the United States.

[E]

We have analyzed the 100 largest companies, the Standard Poor's 100 companies. We've looked at three years, and we have used operating cash flow. That's how investors typically value companies. And we have looked at total shareholder return, and we compared that to the pay that the CEO received.

[A]

So you've looked at pay, you've looked at performance. What do you find?

[E]

I found absolutely no connection between pay and performance. If there was a connection on a scale from 1 to 10, the connection is less than 1.

[A]

Okay, so that's the situation in the U.S. have you looked at other countries?

[E]

Yes, we've looked at other countries. We've looked at Switzerland, we've looked at Germany, and also we looked at the UK and we've just completed that analysis on the FTSE 100 companies, and we have the exact same result as in

[A]

the us so in none of these countries is there any significant correlation between pay and performance.

[E]

There is a little bit better correlation in Germany, actually. Germany has a special situation because they have a much tighter legal framework that holds boards much more accountable for their pay policies.

[A]

So just to defend fatcat chief executives for a moment, if I'm a chief executive of some high performing firm and I'm awarded a bunch of share options in 2005, things are going great. And I sit on these share options and maybe over 2008, 2009, 2010, things haven't gone brilliantly. I'm going to look as though I was paid a huge amount for not doing very much. That might be driving the lack of correlation.

[E]

Not really. When you look at other analyses, they come to the same conclusion. A lot of those compensation components are absolute compensation, which means they reward not just the performance of that company, but actually also the wider economic cycles.

[A]

So you're rewarding a mining company because the price of copper is high. You're rewarding an oil company because the price of oil is high, whereas that's not really a measure of how effectively the company is being managed to.

[E]

Exactly. This is one source for a lack of correlation.

[A]

Herman Stern. And that's all we have time for this week. If you see a fact you'd like us to check out, or a fascinating statistical odyssey which you'd like us to embark upon, drop us an email at more or lessbc.co.uk. our website is bbcworldservice.com More or less the place to subscribe to our podcast. Or you could simply tune in same stat time, same stat channel next week. Goodbye. More or Less was presented by me, Tim Harford, the undercover economist at the Financial Times. The producer was Richard Knight and the editor, Richard Varden. There are dozens of different podcasts now available from the BBC, including news, documentaries, science, business, arts and sport. The details of them all go to bbcworldservice.com podcasts.