Many insurers and reinsurers are following the anti-Government demonstrations in Thailand with significant interest given the heavy financial costs to the industry which followed the previous round of protests in 2010. In this briefing we highlight some relevant issues which (re)insurers may wish to bear in mind as the current situation unfolds.

The recent escalation in political demonstrations in Thailand, particularly in the capital Bangkok, is in many ways reminiscent of the unrest seen in 2010, which ultimately culminated in significant damage to commercial properties in Thailand at a considerable cumulative cost to the (re)insurance industry.

January 2014 has witnessed an escalation in the current unrest, with twin explosions at an anti-Government demonstration in Bangkok in which dozens of people were injured. Overall at least 10 people have now been killed in the disturbances. The property damage seen so far has been relatively minor.

The Thai Government has now declared a state of emergency and, as demonstrations in Bangkok have spread, political analysts  and risk consultancies are warning of further unrest to come. In light of this, many insurers and reinsurers will be concerned as to possible exposure under all risks/property; and terrorism/political violence policies if the current violence continues to escalate.


The latest flare-up in tensions began in November 2013, following an attempt by the Thai Government to pass a political amnesty Bill which would have paved the way for the return of former Prime Minister

Thaksin Shinawatra. Although the Bill was not ultimately approved by the Thai Senate, this move enflamed existing political tensions and led to the heightening of protests against the Government.

In response, Prime Minister Yingluck Shinawatra dissolved Parliament and scheduled a fresh round of elections for early February 2014. However, the opposition Democrat Party declared that it would boycott the vote and made clear that it remains intent on ousting Ms Shinawatra’s government, regardless of whether her Puea Thai Party is re-elected.

The intentions of the military (which has sometimes taken the lead in resolving previous crises) are uncertain, and it is unclear whether the Constitutional Court will play any role in resolving the current situation.

Many of the protest rallies have been held  in central Bangkok, in areas popular within tourists and near major hotels and shopping malls some of which were casualties of the 2010 unrest.

Defining Terrorism and the 2010 unrest

There has historically been no universal consensus on what constitutes “terrorism” for the purposes of insurance contracts in Thailand. As such, following the 2010 unrest, the insurance market was faced with the difficulty of determining whether the property damage which was sustained was caused by “terrorist acts” (as the then government proclaimed), and thereby excluded from cover under standard all-risks policies.

Standard all-risks policy as approved by the Thai Office of Insurance Commission (OIC), provided cover for loss and damage to property unless such damage fell within specified exclusions, including damage caused by acts of “terrorism” or “civil commotion assuming the proportions of or amounting to an uprising” but typically did not define these perils. Terrorism and political violence cover, whether standalone cover or by way of write- back/extension, was taken out by a relatively few businesses.

A number of cases arising out of the 2010 unrest came before the courts (and arbitration tribunals) which were asked to determine whether all-risks property insurers or political violence insurers were liable to indemnity insureds. A central issue was whether the alleged damage to insured property had been caused by acts of “terrorism” or “civil commotion assuming the proportions of or amounting to an uprising (or riot)”.

Despite decisions having been handed down in a number of these cases, questions of liability in future cases are likely to remain contentious, given the widely differing factual circumstances which can lead to specific instances of property damage during periods of unrest.

Evidential Issues

For insurance coverage purposes, each event involving alleged damage must be thoroughly investigated to determine (i) whether the relevant acts fall within the definitions of the terrorism or political violence perils under relevant policy wordings and, if so, (ii)whethertheinsuedorecludd)peril wasapximaecaueofthedama.

Establishing Political Motivation

By its very nature, civil unrest can be hard to categorise, leading to significant evidential difficulties. This is particularly the case when the individuals causing the property damage have not claimed responsibility (or even been arrested) and where their motivations remain unclear.

In particular, proving that specific individuals were motivated by political or ideological purposes when causing damage to insured property can be difficult. A mob which is only interested in plunder and vandalism would not usually be considered to have committed acts of “terrorism” or be part of an “uprising”. The varying motivations of some  of the so-called “black shirts” during the 2010 protests (which had started comparatively peacefully) is a good illustration of this. While political motivations can be relevant, the line between political and economic motivation can often blurred (with economic problems leading to purportedly political actions). Moreover different individuals involved in  one disturbance may have widely different motivations and intentions.

Furthermore, interested parties may seek to put their own gloss on how disturbances are characterised. Pronouncements by incumbent governments on the “terrorist” acts of protestors are unlikely to be determinative for most insurance coverage purposes in the absence of any causative link of these acts  to damage to particular insured property (although for example in the UK, it falls to HM Treasury to confirm that an act of terrorism has occurred for the purposes of cover under the Pool Re scheme).

Insurers are very often hindered by difficulties in obtaining reliable factual evidence. This  can be even more of a problem for reinsurers who are often even further removed. Previous judgments by the Thai courts suggest that  the following types of evidence have been considered in determining the motivations of the perpetrators causing damage to insured property (in no particular order of priority):

  1. Press reports;
  2. CCTV video footage, particularly video  taken at or near the insured property  where the unrest occurred;
  3. Factual witness evidence from employees,  security personnel, police officers, and  reporters, particularly from individuals  who were present at the scene when the  unrest occurred;
  4. Expert witness evidence from forensic fire  experts, in particular, testimony as to the  types of fuel and arms used to start the  fires which caused property damage;
  5. Public statements and speeches from  government officials and protest  leaders. As such, as and when claims  for alleged “riot” or “political violence”  do arise, insurers are advised to obtain  as much evidence as possible, and  as soon as possible. The substance  and timing of public statements and  speeches by Governments, officials and  protest leaders also need to be carefully  examined and weighed (given their  differing agendas).

Thus, where a government alleges that opposition leaders have committed acts of “insurrection” in connection with protests and/or issued arrest warrants on this basis, this is unlikely to be determinative for insurance coverage purposes without specific references to the time and place of the alleged political acts.

Establishing Causation

Proving the peril exists is one step,  establishing that it caused the loss is the  necessary next step.

In practical terms, demonstrating  geographical and temporal proximity between  the peril and the insured property is critical.  It would be difficult for insurers to maintain  that a loss was caused by terrorist acts if the  acts were not geographically proximate to the  location where the loss occurred, or if the loss  did not occur at the time at which the alleged  political acts were taking place.

Thus, where a government alleges that  opposition leaders have committed acts of  “insurrection” in connection with protests  and/or issued arrest warrants on this basis,  this is unlikely to be determinative for  insurance coverage purposes without specific  references to the time and place of the  alleged political acts.

In Pan American World Airways Inc v Aetna  Casualty & Surety Co 1 , where the policy  excluded liability for “loss or damage due  to or resulting from… civil commotion”, the  court found that a hijacking over London  of an airplane (the “loss”) that never went  or was intended to go to Jordan could  not be deemed to be the result of a civil  commotion occurring in Jordan.Displacing  proximate cause.

Tests for causation may themselves be altered  by specific wording, by phrases seeking to  circumvent the normal rules on proximate  cause (and widen the relevant exclusion).

Some clauses may seek to dilute the  requirement of a causative link between the  loss and the peril by stating that the terrorism  exclusion only applies if “directly or indirectly  caused by” terrorism. Under English law,  the words “indirectly caused by” indicates  that the proximate cause requirement does  not strictly apply. That said, some causal  connection, which is not too remote, must  still be established.

Reversing the burden of Proof

Furthermore, “reverse burden of proof”  provisions are found in many standard  terrorism exclusions in Thai policies. The  intention of such provisions, as the words  suggest, is to reverse the normal onus of  proof by requiring an insured to disprove  applicability of an exclusion. This may  state: “If Insurers allege that by reason of  this exclusion, any loss, damage, cost or  expense is not covered by this insurance the  burden of proving the contrary shall be upon  the Insured”.

The validity of such provisions has been  recognised by English Courts. However, they  have also ruled that insurers must still make  out a prima facie case that the loss was caused  by an excluded peril.Punitive Damages for  Denial or Late Payment of Claims

Another possible concern for insurers is the  possibility of liability to pay punitive damages  for the denial or late payment of a claim.  Under the Thai Consumer Case Procedure  Act 2008 (which applies to both individuals  and corporations), punitive damages may  be awarded against insurers if payments are  wilfully withheld or delayed in bad faith. This  Act was invoked in cases arising from the  previous round of protests in 2010.

Insurers need to balance the need to  properly investigate claims against the  disproportionate economic consequences of  possible punitive damages award. Thankfully  there are often a number of practical solutions  to these types of problem.

Due to changes in market wordings, some of  the issues faced by the insurers and reinsurers  following the 2010 protests may be avoided  if events repeat themselves. In particular,  many political violence policies now cover  a much broader range of perils including  cover for malicious damage; riots, strikes and  civil commotion.

However where insureds do not have  comprehensive political violence cover  and suffer losses at times of unrest, they  will inevitably look to see what advice and  options were provided by their brokers prior  to placement.


Several years have passed since the last  major round of protests in Bangkok in 2010.  However once again the industry is closely  watching developments unfold. Tensions in  many other countries such as the Ukraine,  Egypt, Turkey and Libya are also currently in  the spotlight.Even with more widespread use  of surveillance technology and advances in  forensic science, when claims arise, evidential  challenges will remain in one form or another.

Looking ahead, a continuing re-examination  of wordings may be necessary to ensure  that policies clearly define and differentiate  between insured and excluded risks, and  clearly identify the causal connection required  between the excluded event and the loss  in question.