# The string that binds currencies

Exchange Rate Series: Topic 2

In theory, there are known links between currency groups. Some nations peg their currency to others, Some are closely linked with commodities such as Oil prices.

We have the tools and the curiosity, so why not attempt to investigate and derive some of these links with datasets available to us?

Note: All currency exchange rate values represented use Singapore dollars ad the base(S\$).

# Currency pegs

In theory, some countries peg their currency value to others, there are multiple reasons for this, but can we identify some links with data we have?

## Correlation between currency pairs

Correlation helps us to understand how ‘related’ 2 variables are. Below we find a very colorful representation of the different combinations of USD, AUD, EUR, GBP, MYR, JPY & RMB across 2002–2021.

From these plots we can already see some relationships, for example, MYR & JPY show a positive correlation between them in the year 2021.

But let’s look at another pair which appear to be long term dance partners,

S\$/USD across the years shows a very strong positive correlation with S\$/RMB. We could make an informed guess here that one of them has pegged itself to the other, but from this data, we are unable to determine who is a dance partner who wouldn’t let go. No prizes for guessing though.

Investopedia confirms our hypothesis and informs us that the RMB is pegged to the USD. This allows China’s exports to remain competitive in the US (preventing appreciation of the Yuan). There are key economic terms such as floating exchange rates to explore, but let’s not go into those economics here.

The above article gives us one more concept to explore,

How tightly were they holding on to each other?

## The Break Up

2018 appears as an interesting data point, that year we observe the first and largest negative correlation. Doesn’t take long to draw the link, but the effect of the trade war between the countries can be seen here!

## The loosening grip

Let us use linear regression without a dataset to model both currencies with a straight line. Where X = amount of USD per SGD and Y = amount of RMB per SGD.

Thus, RMB = (gradient) * USD + (intercept)

# Wonderful looks like we were able to get similar results to what is mentioned in theory!

From the earlier Investopedia article, here are some key events

The yuan was pegged to the greenback at 8.28 to the dollar.

• In our dataset, we successfully recovered the 8.28 to the dollar before 2005!

July 2005, the yuan was permitted to appreciate by 2.1% against the dollar.

• In our dataset, we notice an appreciation after 2005!

2005–2008 the yuan was allowed to appreciate to a level of 6.83 to the dollar.

• In 2009 we observe a value of 6.75 to the dollar!

I did not report intercept values as for the key points the intercept amount was negligible, however, a strong bias creeps in later years.

# So what have we learned?

1. With curiosity and enthusiasm, we can derive insights found in theory adn articles using real datasets. Be your own fact checker!
2. We can visualize how currency value is related to other factors such as foreign policy and commodity prices such as oil.
3. Understanding these relationships can help us make better decisions on currency pairs to invest in or hold on to… but that is not for me to advise, maybe next time.