Merge pull request SOMAS2020#65 from ezgimez/ostrom-2

[IIGO] [2/3] More Analysis and Evaluation

05_iigo/index.tex

@@ -975,36 +975,56 @@ \subsubsection{Institutional Analysis and Development framework overview of IIGO
 
         \item The third decision arena of IIGO is that surrounding sanctions. Here the Judge is the only actor involved in the decision making of whether to and how harshly to sanction islands, as well as whether to pardon islands.
 
-        \item The fourth decision arena of IIGO is that surrounding monitoring of role powers and elections. The two are interlinked whereby the elected actors have to participate in not only the decision making surrounding elections but also the decision making in monitoring, which in turn can alter the rules in play to oblige other actors to hold elections. Moreover, elections involve the involvement of all islands voting, thus providing information to the island responsible for the appointment of a role. It is important to note the repetitive nature of decisions roles have to make regarding elections, followed by decisions regarding monitoring, followed by elections, and so on. 
-        
+        \item The fourth decision arena of IIGO is that surrounding monitoring of role powers and elections. The two are interlinked whereby the elected actors have to participate in not only the decision making surrounding elections but also the decision making in monitoring, which in turn can alter the rules in play to oblige other actors to hold elections. Moreover, elections involve the involvement of all voting islands, thus providing information to the island responsible for the appointment of a role. It is important to note the repetitive nature of decisions roles have to make regarding elections, followed by decisions regarding monitoring, followed by elections, and so on.
+
         \item The fifth decision arena involves the roles each of the IIGO roles decision regarding managing the costs of IIGO actions. For each action with a cost, the only actor involved is the agent holding the power. There is, however, a budget rule, that grants permission to act on these powers only if there is enough budget. The budget is incremented every turn and the amount of incrementation is to be set by the decision group described in the second decision arena.
 
         \item The sixth decision arena involves the roles each of the IIGO roles decision regarding managing the salary of other IIGO roles. Similarly to the budget, the only actor involved in this decision is the role that holds the power to make the transaction - a flow of resources from the common pool to the respective (see Figure~\ref{fig:cycles_in_IIGO}) islands resource pool. This decision is affected by the salary rule (Rule~\ref{rule:salary}) which just like with budget is set by the decision group described in the second decision arena.
 
-        \item The seventh and final action arena IIGO involves the actual appropriation and provision of resources all agents perform each round. Here the decision is in the hands of each island, however, the rules surrounding the decision are formed by the decision group described in the first decision arena. 
+        \item The seventh and final Action Arena IIGO involves the actual appropriation and provision of resources all agents perform each round. Here the decision is in the hands of each island, however, the rules surrounding the decision are formed by the decision group described in the first decision arena. 
     \end{enumerate}
 
 \end{itemize}
 \subsubsection {Evaluation of IIGO with regard to Ostrom’s Design Principles} \label{sec:ostrom_eval}
 \begin{enumerate}
     \item Clearly defined boundaries: those who have rights or entitlement to appropriate resources from the common pool are clearly defined, as are its boundaries.
 
-    To evaluate this principle we must look at action arenas 1 and 7. The seventh action arena tells us that every agent decides as to whether to appropriate resources. The first tells us that the President can enable clear boundaries by providing allocations.
+    To evaluate this principle we must look at Action Arenas 1 and 7. The seventh Action Arena tells us that every agent decides as to whether to appropriate resources. The first tells us that the President can enable clear boundaries by providing allocations.
 
     \item Congruence between appropriation and provision rules and the state of the prevailing local environment.
 
-    To evaluate this principle we must look at action arena 1. We can quickly realise that ensuring congruence is a decision of the President. Hence there is an outcome of the system where an agent that is not concerned with the welfare of others or the long term common pool risk dilemma might set out taxes larger than the islands can even afford. To solve this problem we can introduce a Zone of Dignity. This would involve defining metrics and rules that state how the President's policies must measure up to the metrics. These metrics and rules, such as an islands resource amount below which a President is not permitted to tax an island, could be set by the legislative branch.
+    To evaluate this principle we must look at Action Arena 1. We can quickly realise that ensuring congruence is a decision of the President. Hence there is an outcome of the system where an agent that is not concerned with the welfare of others or the long term common pool risk dilemma might set out taxes larger than the islands can even afford. To solve this problem we can introduce a Zone of Dignity. This would involve defining metrics and rules that state how the President's policies must measure up to the metrics. These metrics and rules, such as an islands resource amount below which a President is not permitted to tax an island, could be set by the legislative branch.
 
     \item Collective-choice arrangements: in particular, self-determination, whereby those affected by the rules participate in the selection, modification and enforcement of those rules.
 
-    To evaluate this principle we must look at action arena 2 and 3. The selection of rules is related to the process of choosing a rule to vote on - islands decide and propose rules for the President to select one. We can notice that the institution is lacking. Which rule to select is a decision made by the President. Moreover, the President also proposes rules the same as every other agent. Hence, while all those affected participate by providing their desired rules to the President, that information may not be utilised. It is possible for opinion formation and the system of elections to, at least in part, remedy this. An agent would be less likely to vote for a President which selects rules that are distant from the agent's optimal rule-set. In an institutional state of oligarchy, this system would have no effect, so another solution might be needed. 
+    To evaluate this principle we must look at Action Arena 2 and 3. The selection of rules is related to the process of choosing a rule to vote on - islands decide and propose rules for the President to select one. We can notice that the institution is lacking. Which rule to select is a decision made by the President. Moreover, the President also proposes rules the same as every other agent. Hence, while all those affected participate by providing their desired rules to the President, that information may not be utilised. It is possible for opinion formation and the system of elections to, at least in part, remedy this. An agent would be less likely to vote for a President which selects rules that are distant from the agent's optimal rule-set. In an institutional state of oligarchy, this system would have no effect, so another solution might be needed. 
 
     Rule modification is the result of the voting system. If all the available rules governing this process are followed by the agents, all agents equally contribute to this decision.  
 
     Of note here is the Speaker's power to decide which islands are to participate in the vote. This power is governed by Rule~\ref*{rule:all_islands_vote} and is a remnant of a system of sanctions we considered during design, which could bar islands from participating in a vote. It is the Speakers power to call a vote, i.e. send a communication message to all the agents, hence it would have been the Speakers power to enforce such sanctions, by choosing which islands to send a vote request to. This system is not implemented, which begs for such a decision to be more regulated through a constitutional rule.
 
-    The enforcement of collective choice rules is covered by Ostrom's 4th principle.
+
+    The enforcement of collective choice rules is discussed in with Ostrom's 4th principle.
+
+    \item Monitoring, of both state conditions and appropriator behaviour, is by appointed agencies, who are either accountable to the resource appropriators or are appropriators themselves.
 
+    The first set of monitoring is performed by the Judge. It is described in Action Arena 3. While the Judge is always a resource appropriator, under an institutional state of tyranny or oligarchy, where the role of the Judge is only ever in the hands of one or a few islands, the decision group consists of only one island. Moreover, in the current IIGO implementation, there is no way to monitor the behaviour of the Judge in this regard, there is no way to confirm that the monitoring results announced by the Judge are concurrent with the state and events of the game. This is an issue that should be addressed in future iterations of IIGO.
+
+    The second set of monitoring is that which roles perform in what is referred to as the Accountability Cycle (see Section~\ref{sec:accountability}). Whether roles abide by the collective choice rules in play is monitored by the roles themselves. The most notable flaw in this section is the one that was alluded to in the repetitive nature of elections and monitoring. There is a possible outcome of the institution where the agent responsible for the appointment disregards the election process and appoints themselves. The reverse nature of the transfer-of-power and monitoring cycles means that the same agent is then responsible for monitoring their own inappropriate behaviour. Within one degree of monitoring, where monitoring itself is not monitored, the agent would be accountable to only itself. 
+
+    It can be argued that a log of the decisions made by roles in power should be public information. Combined with a system to democratically remove roles, that is, to hold elections that are hosted by any agent, this form of public accountability could be a potential solution. Another solution would be to introduce a system of appealing the result of an election, whereby an election result is not recognised until it is acknowledged as truthful by the monitoring agency. 
+
+
+    \item A flexible scale of graduated sanctions for resource appropriators who violate communal rules; i.e. violation should be proportional to the severity, frequency or necessity of the offence.
+
+    While the agents are given a flexible system of graduated sanctions to utilise when in the role of the Judge in the form of sanction tiers and setting the severity of offence, these decisions are entirely and freely made by the Judge. There are no institutional "guardrails", there is no Zone of Dignity for this process. An agent could choose to not sanction itself or to not sanction severe and frequent offences while still heavily sanctioning agents which are minor one-time offenders. While the sanctioning mechanism is present, the lack of a standard protocol indicates that the current implementation of IIGO is lacking with regard to Ostrom's fifth principle.
+
+    When looked at the sanctions related to the offences that roles can make when in power, IIGO is also lacking in this regard. The worst punishment any agent can receive for not abiding by the rules that govern the decision-making in the 3 government branches is that the agent can be let go from the role prematurely. Hence the only thing an agent can lose is the potential for future salary. Hence, there are situations where agents could calculate that the optimal strategy is to always abuse these powers for their own benefit until they are caught.
+
+    \item Universal access to ‘fair’, rapid, low-cost, light-weight conflict-resolution mechanisms.
+
+    When looking at the current implementation of IIGO we can notice that monitoring and enforcement are heavily linked and both decisions for operational-rule offences are held by the Judge. A sanctioned island would have no way of appealing the decision made by the Judge. Hence, there is no conflict-resolution mechanism. Moreover, there is no "innocent until proven guilty" mechanism. One way to remedy this, without introducing a Zone of Dignity for the Judge's decisions, which can then be monitored. An alternative could be to introduce an Alternative Dispute Resolution protocol, with respective meta-rules, as well as the idea of a court in cases where the Alternative Dispute Resolution would necessitate litigation. 
+
 \end{enumerate}
 \subsection{Future Work}